Datasets:

andersonbcdefg

/

arxiv-triples-filtered

like 1

2004.08168

A Bayesian comparison between $\Lambda$CDM and phenomenologically emergent dark energy models

In this work we examine the recently proposed phenomenological emergent dark energy (PEDE) model by \cite{Li:2019yem}, using the latest observational data in both expansion and perturbation levels. Applying the statistical Bayesian evidence as well as the AIC and BIC information criteria, we compare the PEDE model with the concordance $\Lambda$CDM model in both flat and non-flat universes. We combine the observational datasets as (i) expansion data (except CMB), (ii) expansion data (including CMB) and (iii) expansion data jointed to the growth rate dataset. Our statistical results show that the flat- $\Lambda$CDM model is still the best model. In the case of expansion data (including CMB), we observe that the flat- PEDE model is well consistent with observations as well as the concordance $\Lambda$CDM universe. While in the cases of (i) and (iii), the PEDE models in both of the flat and non-flat geometries are not favored. In particular, we see that in the perturbation level the PEDE model can not fit the observations as equally as standard $\Lambda$CDM cosmology. As the ability of the model, we show that the PEDE models can alleviate the tension of Hubble constant value appearing between the local observations and Planck inferred estimation in standard cosmology.

We test a class of holographic models for the very early universe against cosmological observations and find that they are competitive to the standard $\Lambda$CDM model of cosmology. These models are based on three dimensional perturbative super-renormalizable Quantum Field Theory (QFT), and while they predict a different power spectrum from the standard power-law used in $\Lambda$CDM, they still provide an excellent fit to data (within their regime of validity). By comparing the Bayesian evidence for the models, we find that $\Lambda$CDM does a better job globally, while the holographic models provide a (marginally) better fit to data without very low multipoles (i.e. $l\lesssim 30$), where the dual QFT becomes non-perturbative. Observations can be used to exclude some QFT models, while we also find models satisfying all phenomenological constraints: the data rules out the dual theory being Yang-Mills theory coupled to fermions only, but allows for Yang-Mills theory coupled to non-minimal scalars with quartic interactions. Lattice simulations of 3d QFT's can provide non-perturbative predictions for large-angle statistics of the cosmic microwave background, and potentially explain its apparent anomalies.

2004.08169

Splitting type variational problems with linear growth conditions

Regularity properties of solutions to variational problems are established for a broad class of strictly convex splitting type energy densities of the principal form $f$: $\mathbb{R}^2 \to \mathbb{R}$, \[ f(\xi_1,\xi_2) = f_1\big( \xi_1 \big) + f_2\big( \xi_2 \big) \, , \] with linear growth. As a main result it is shown that, regardless of a corresponding property of $f_2$, the assumption ($t\in \mathbb{R}$) \[ c_1 (1+|t|)^{-\mu_{1}} \le f_1(t) \le c_2\, ,\quad 1 \le \mu_1 < 2\, , \] is sufficient to obtain higher integrability of $\partial_1 u$ for any finite exponent. Similar results in the case $f$: $\mathbb{R}^n \to \mathbb{R}$ hold with the obvious changes in notation.

This is the first of a sequence of papers proving the quantum invariance under ordinary flops over an arbitrary smooth base. In this first part, we determine the defect of the cup product under the canonical correspondence and show that it is corrected by the small quantum product attached to the extremal ray. We then perform various reductions to reduce the problem to the local models. In Part II, we develop a quantum Leray--Hirsch theorem and use it to show that the big quantum cohomology ring is invariant under analytic continuations in the K\"ahler moduli space for ordinary flops of splitting type. In Part III, together with F. Qu, we remove the splitting condition by developing a quantum splitting principle, and hence solve the problem completely.

2004.08170

Deep Echo State Networks for Short-Term Traffic Forecasting: Performance Comparison and Statistical Assessment

In short-term traffic forecasting, the goal is to accurately predict future values of a traffic parameter of interest occurring shortly after the prediction is queried. The activity reported in this long-standing research field has been lately dominated by different Deep Learning approaches, yielding overly complex forecasting models that in general achieve accuracy gains of questionable practical utility. In this work we elaborate on the performance of Deep Echo State Networks for this particular task. The efficient learning algorithm and simpler parametric configuration of these alternative modeling approaches make them emerge as a competitive traffic forecasting method for real ITS applications deployed in devices and systems with stringently limited computational resources. An extensive comparison benchmark is designed with real traffic data captured over the city of Madrid (Spain), amounting to more than 130 automatic Traffic Readers (ATRs) and several shallow learning, ensembles and Deep Learning models. Results from this comparison benchmark and the analysis of the statistical significance of the reported performance gaps are decisive: Deep Echo State Networks achieve more accurate traffic forecasts than the rest of considered modeling counterparts.

Air quality forecasting has been regarded as the key problem of air pollution early warning and control management. In this paper, we propose a novel deep learning model for air quality (mainly PM2.5) forecasting, which learns the spatial-temporal correlation features and interdependence of multivariate air quality related time series data by hybrid deep learning architecture. Due to the nonlinear and dynamic characteristics of multivariate air quality time series data, the base modules of our model include one-dimensional Convolutional Neural Networks (1D-CNNs) and Bi-directional Long Short-term Memory networks (Bi-LSTM). The former is to extract the local trend features and spatial correlation features, and the latter is to learn spatial-temporal dependencies. Then we design a jointly hybrid deep learning framework based on one-dimensional CNNs and Bi-LSTM for shared representation features learning of multivariate air quality related time series data. We conduct extensive experimental evaluations using two real-world datasets, and the results show that our model is capable of dealing with PM2.5 air pollution forecasting with satisfied accuracy.

2004.08171

Introducing the HD+B model for pulsar wind nebulae: a hybrid hydrodynamics/radiative approach

Identification and characterization of a rapidly increasing number of pulsar wind nebulae is, and will continue to be, a challenge of high-energy gamma-ray astrophysics. Given that such systems constitute -- by far -- the most numerous expected population in the TeV regime, such characterization is important not only to learn about the sources per se from an individual and population perspective, but also to be able to connect them with observations at other frequencies, especially in radio and X-rays. Also, we need to remove the emission from nebulae in highly confused regions of the sky for revealing other underlying emitters. In this paper we present a new approach for theoretical modelling of pulsar wind nebulae: a hybrid hydrodynamic-radiative model able to reproduce morphological features and spectra of the sources, with relatively limited numerical cost.

Motivated by recent detections of pulsar wind nebulae in very-high-energy (VHE) gamma rays, a systematic search for VHE gamma-ray sources associated with energetic pulsars was performed, using data obtained with the H.E.S.S. (High Energy Stereoscopic System) instrument. The search for VHE gamma-ray sources near the pulsar PSR J1718-3825 revealed the new VHE gamma-ray source HESS J1718-385. We report on the results from the HESS data analysis of this source and on possible associations with the pulsar and at other wavelengths. We investigate the energy spectrum of HESS J1718-385 that shows a clear peak. This is only the second time a VHE gamma-ray spectral maximum from a cosmic source was observed, the first being the Vela X pulsar wind nebula.

2004.08172

Finding the Optimal Network Depth in Classification Tasks

We develop a fast end-to-end method for training lightweight neural networks using multiple classifier heads. By allowing the model to determine the importance of each head and rewarding the choice of a single shallow classifier, we are able to detect and remove unneeded components of the network. This operation, which can be seen as finding the optimal depth of the model, significantly reduces the number of parameters and accelerates inference across different hardware processing units, which is not the case for many standard pruning methods. We show the performance of our method on multiple network architectures and datasets, analyze its optimization properties, and conduct ablation studies.

Heterogeneous data modalities can provide complementary cues for several tasks, usually leading to more robust algorithms and better performance. However, while training data can be accurately collected to include a variety of sensory modalities, it is often the case that not all of them are available in real life (testing) scenarios, where a model has to be deployed. This raises the challenge of how to extract information from multimodal data in the training stage, in a form that can be exploited at test time, considering limitations such as noisy or missing modalities. This paper presents a new approach in this direction for RGB-D vision tasks, developed within the adversarial learning and privileged information frameworks. We consider the practical case of learning representations from depth and RGB videos, while relying only on RGB data at test time. We propose a new approach to train a hallucination network that learns to distill depth information via adversarial learning, resulting in a clean approach without several losses to balance or hyperparameters. We report state-of-the-art results on object classification on the NYUD dataset and video action recognition on the largest multimodal dataset available for this task, the NTU RGB+D, as well as on the Northwestern-UCLA.

2004.08173

On open books for nonorientable 3-manifolds

We show that the monodromy of Klassen's genus two open book for $P^2 \times S^1$ is the $Y$-homeomorphism of Lickorish, which is also known as the crosscap slide. Similarly, we show that $S^2 \widetilde{\times} S^1$ admits a genus two open book whose monodromy is the crosscap transposition. Moreover, we show that each of $P^2 \times S^1$ and $S^2 \widetilde{\times} S^1$ admits infinitely many isomorphic genus two open books whose monodromies are mutually nonisotopic. Furthermore, we include a simple observation about the stable equivalence classes of open books for $P^2 \times S^1$ and $S^2 \widetilde{\times} S^1$. Finally, we formulate a version of Gabai's theorem about the Murasugi sum of open books, without imposing any orientability assumption on the pages.

The book deals with a stochastic formulation of path integration in real time, by rotating the_space_ variables over exp(i pi/4). Preliminary chapters deal with quantum and classical mechanics, probability theory and stochastic calculus, and the conventional approach to path integration through the Trotter product formula. Subsequently, the stochastic formulation in complex space is established, and applied to semi-classical expansions, nonlinear oscillations, quantum dynamics on analytic manifolds, dissipative systems (i.e., those which are open in that they interact with an environment), tunneling, field theory, and computer simulations.

2004.08175

Nano- and microscale apertures in metal films fabricated by colloidal lithography with perovskite nanocrystals

We demonstrate patterning of metal surfaces based on lift-off of perovskite nanocrystals that enables the fabrication of nanometer-size features without the use of resist-based nanolithography. The perovskite nanocrystals act as templates for defining the shape of the apertures in metal layers, and we exploit the variety of sizes and shapes that can be controlled in the colloidal synthesis to demonstrate the fabrication of nanoholes, nanogaps and guides with size smaller than the wavelength of light in the visible spectrum. The process can be readily integrated with standard lithography and etching techniques for the creation of more complex structures.

Nano granular metallic iron (Fe) and titanium dioxide (TiO$_{2-\delta}$) were co-deposited on (100) lanthanum aluminate (LaAlO$_3$) substrates in a low oxygen chamber pressure using a pulsed laser ablation deposition (PLD) technique. The co-deposition of Fe and TiO$_2$ resulted in $\approx$ 10 nm metallic Fe spherical grains suspended within a TiO$_{2-\delta}$ matrix. The films show ferromagnetic behavior with a saturation magnetization of 3100 Gauss at room temperature. Our estimate of the saturation magnetization based on the size and distribution of the Fe spheres agreed well with the measured value. The film composite structure was characterized as p-type magnetic semiconductor at 300 K with a carrier density of the order of $ 10^{22} /{\rm cm^3}$. The hole carriers were excited at the interface between the nano granular Fe and TiO$_{2-\delta}$ matrix similar to holes excited in the metal/n-type semiconductor interface commonly observed in Metal-Oxide-Semiconductor (MOS) devices. From the large anomalous Hall effect directly observed in these films it follows that the holes at the interface were strongly spin polarized. Structure and magneto transport properties suggested that these PLD films have potential nano spintronics applications.

2004.08176

Exploring time-series motifs through DTW-SOM

Motif discovery is a fundamental step in data mining tasks for time-series data such as clustering, classification and anomaly detection. Even though many papers have addressed the problem of how to find motifs in time-series by proposing new motif discovery algorithms, not much work has been done on the exploration of the motifs extracted by these algorithms. In this paper, we argue that visually exploring time-series motifs computed by motif discovery algorithms can be useful to understand and debug results. To explore the output of motif discovery algorithms, we propose the use of an adapted Self-Organizing Map, the DTW-SOM, on the list of motif's centers. In short, DTW-SOM is a vanilla Self-Organizing Map with three main differences, namely (1) the use the Dynamic Time Warping distance instead of the Euclidean distance, (2) the adoption of two new network initialization routines (a random sample initialization and an anchor initialization) and (3) the adjustment of the Adaptation phase of the training to work with variable-length time-series sequences. We test DTW-SOM in a synthetic motif dataset and two real time-series datasets from the UCR Time Series Classification Archive. After an exploration of results, we conclude that DTW-SOM is capable of extracting relevant information from a set of motifs and display it in a visualization that is space-efficient.

Transcriptional regulation by transcription factors and post-transcriptional regulation by microRNAs constitute two major modes of regulation of gene expression. While gene expression motifs incorporating solely transcriptional regulation are well investigated, the dynamics of motifs with dual strategies of regulation, i.e., both transcriptional and post-transcriptional regulation, have not been studied as extensively. In this paper, we probe the dynamics of a four-gene motif with dual strategies of regulation of gene expression. Some of the functional characteristics are compared with those of a two-gene motif, the genetic toggle, employing only transcriptional regulation. Both the motifs define positive feedback loops with the potential for bistability and hysteresis. The four-gene motif, contrary to the genetic toggle, is found to exhibit bistability even in the absence of cooperativity in the regulation of gene expression. The four-gene motif further exhibits a novel dynamical feature in which two regions of monostability with linear threshold response are separated by a region of bistability with digital response. Using the linear noise approximation (LNA), we further show that the coefficient of variation (a measure of noise), associated with the protein levels in the steady state, has a lower magnitude in the case of the four-gene motif as compared to the case of the genetic toggle. We next compare transcriptional with post-transcriptional regulation from an information theoretic perspective. We focus on two gene expression motifs, Motif 1 with transcriptional regulation and Motif 2 with post-transcriptional regulation. We show that amongst the two motifs, Motif 2 has a greater capacity for information transmission for an extended range of parameter values.

2004.08177

A Data-Driven Frequency Scaling Approach for Deadline-aware Energy Efficient Scheduling on Graphics Processing Units (GPUs)

Modern computing paradigms, such as cloud computing, are increasingly adopting GPUs to boost their computing capabilities primarily due to the heterogeneous nature of AI/ML/deep learning workloads. However, the energy consumption of GPUs is a critical problem. Dynamic Voltage Frequency Scaling (DVFS) is a widely used technique to reduce the dynamic power of GPUs. Yet, configuring the optimal clock frequency for essential performance requirements is a non-trivial task due to the complex nonlinear relationship between the application's runtime performance characteristics, energy, and execution time. It becomes more challenging when different applications behave distinctively with similar clock settings. Simple analytical solutions and standard GPU frequency scaling heuristics fail to capture these intricacies and scale the frequencies appropriately. In this regard, we propose a data-driven frequency scaling technique by predicting the power and execution time of a given application over different clock settings. We collect the data from application profiling and train the models to predict the outcome accurately. The proposed solution is generic and can be easily extended to different kinds of workloads and GPU architectures. Furthermore, using this frequency scaling by prediction models, we present a deadline-aware application scheduling algorithm to reduce energy consumption while simultaneously meeting their deadlines. We conduct real extensive experiments on NVIDIA GPUs using several benchmark applications. The experiment results have shown that our prediction models have high accuracy with the average RMSE values of 0.38 and 0.05 for energy and time prediction, respectively. Also, the scheduling algorithm consumes 15.07% less energy as compared to the baseline policies.

We report an efficient algorithm for calculating momentum-space integrals in solid state systems on modern graphics processing units (GPUs). Our algorithm is based on the tetrahedron method, which we demonstrate to be ideally suited for execution in a GPU framework. In order to achieve maximum performance, all floating point operations are executed in single precision. For benchmarking our implementation within the CUDA programming framework we calculate the orbital-resolved density of states in an iron-based superconductor. However, our algorithm is general enough for the achieved improvements to carry over to the calculation of other momentum integrals such as, e.g. susceptibilities. If our program code is integrated into an existing program for the central processing unit (CPU), i.e. when data transfer overheads exist, speedups of up to a factor $\sim130$ compared to a pure CPU implementation can be achieved, largely depending on the problem size. In case our program code is integrated into an existing GPU program, speedups over a CPU implementation of up to a factor $\sim165$ are possible, even for moderately sized workloads.

2004.08178

Highway Transformer: Self-Gating Enhanced Self-Attentive Networks

Self-attention mechanisms have made striking state-of-the-art (SOTA) progress in various sequence learning tasks, standing on the multi-headed dot product attention by attending to all the global contexts at different locations. Through a pseudo information highway, we introduce a gated component self-dependency units (SDU) that incorporates LSTM-styled gating units to replenish internal semantic importance within the multi-dimensional latent space of individual representations. The subsidiary content-based SDU gates allow for the information flow of modulated latent embeddings through skipped connections, leading to a clear margin of convergence speed with gradient descent algorithms. We may unveil the role of gating mechanism to aid in the context-based Transformer modules, with hypothesizing that SDU gates, especially on shallow layers, could push it faster to step towards suboptimal points during the optimization process.

The medical image is characterized by the inter-class indistinction, high variability, and noise, where the recognition of pixels is challenging. Unlike previous self-attention based methods that capture context information from one level, we reformulate the self-attention mechanism from the view of the high-order graph and propose a novel method, namely Hierarchical Attention Network (HANet), to address the problem of medical image segmentation. Concretely, an HA module embedded in the HANet captures context information from neighbors of multiple levels, where these neighbors are extracted from the high-order graph. In the high-order graph, there will be an edge between two nodes only if the correlation between them is high enough, which naturally reduces the noisy attention information caused by the inter-class indistinction. The proposed HA module is robust to the variance of input and can be flexibly inserted into the existing convolution neural networks. We conduct experiments on three medical image segmentation tasks including optic disc/cup segmentation, blood vessel segmentation, and lung segmentation. Extensive results show our method is more effective and robust than the existing state-of-the-art methods.

2004.08179

Core/Shell CdSe/CdS Bone-Shaped Nanocrystals with a Thick and Anisotropic Shell as Optical Emitters

Colloidal core/shell nanocrystals are key materials for optoelectronics, enabling control over essential properties via precise engineering of the shape, thickness, and crystal lattice structure of their shell. Here, we apply the growth protocol for CdS branched nanocrystals on CdSe nanoplatelet seeds and obtain bone-shaped heterostructures with a highly anisotropic shell. Surprisingly, the nanoplatelets withstand the high growth temperature of 350 {\deg}C and we obtain structures with a CdSe nanoplatelet core that is overcoated by a shell of cubic CdS, on top of which tetrahedral CdS structures with hexagonal lattice are formed. These complex core/shell nanocrystals show a bandedge emission around 657 nm with a photoluminescence quantum yield of ca. 42 % in solution, which is also retained in thin films. Interestingly, the nanocrystals manifest simultaneous red and green emission, and the relatively long wavelength of the green emission indicates charge recombination at the cubic/hexagonal interface of the CdS shell. The nanocrystal films show amplified spontaneous emission, random lasing, and distributed feedback lasing when the material is deposited on suitable gratings. Our work stimulates the design and fabrication of more exotic core/shell heterostructures where charge carrier delocalization, dipole moment, and other optical and electrical properties can be engineered.

Achieving significant doping in GaAs/AlAs core/shell nanowires (NWs) is of considerable technological importance but remains a challenge due to the amphoteric behavior of the dopant atoms. Here we show that placing a narrow GaAs quantum well in the AlAs shell effectively getters residual carbon acceptors leading to an \emph{unintentional} p-type doping. Magneto-optical studies of such a GaAs/AlAs core multi-shell NW reveal quantum confined emission. Theoretical calculations of NW electronic structure confirm quantum confinement of carriers at the core/shell interface due to the presence of ionized carbon acceptors in the 1~nm GaAs layer in the shell. Micro-photoluminescence in high magnetic field shows a clear signature of avoided crossings of the $n=0$ Landau level emission line with the $n=2$ Landau level TO phonon replica. The coupling is caused by the resonant hole-phonon interaction, which points to a large 2D hole density in the structure.

2004.08180

A Hierarchical Convex Optimization for Multiclass SVM Achieving Maximum Pairwise Margins with Least Empirical Hinge-Loss

In this paper, we formulate newly a hierarchical convex optimization for multiclass SVM achieving maximum pairwise margins with least empirical hinge-loss. This optimization problem is a most faithful as well as robust multiclass extension of an NP-hard hierarchical optimization appeared for the first time in the seminal paper by C.~Cortes and V.~Vapnik almost 25 years ago. By extending the very recent fixed point theoretic idea [Yamada-Yamagishi 2019] with the generalized hinge loss function [Crammer-Singer 2001], we show that the hybrid steepest descent method [Yamada 2001] in the computational fixed point theory is applicable to this much more complex hierarchical convex optimization problem.

An ideal description for a given video should fix its gaze on salient and representative content, which is capable of distinguishing this video from others. However, the distribution of different words is unbalanced in video captioning datasets, where distinctive words for describing video-specific salient objects are far less than common words such as 'a' 'the' and 'person'. The dataset bias often results in recognition error or detail deficiency of salient but unusual objects. To address this issue, we propose a novel learning strategy called Information Loss, which focuses on the relationship between the video-specific visual content and corresponding representative words. Moreover, a framework with hierarchical visual representations and an optimized hierarchical attention mechanism is established to capture the most salient spatial-temporal visual information, which fully exploits the potential strength of the proposed learning strategy. Extensive experiments demonstrate that the ingenious guidance strategy together with the optimized architecture outperforms state-of-the-art video captioning methods on MSVD with CIDEr score 87.5, and achieves superior CIDEr score 47.7 on MSR-VTT. We also show that our Information Loss is generic which improves various models by significant margins.

2004.08181

A colored Khovanov bicomplex

In this note, we prove the existence of a tri-graded Khovanov-type bicomplex (Theorem 1.2). The graded Euler characteristic of the total complex associated with this bicomplex is the colored Jones polynomial of a link. The first grading of the bicomplex is a homological one derived from cabling of the link (i.e., replacing a strand of the link with several parallel strands); the second grading is related to the homological grading of ordinary Khovanov homology; finally, the third grading is preserved by the differentials, and corresponds to the degree of the variable in the colored Jones polynomial. In particular, we introduce a way to take a small cabling link diagram directly from a big cabling link diagram (Theorem 3.2). This is a refined version of arXiv:0907.5247.

Quasi-alternating links are homologically thin for both Khovanov homology and knot Floer homology. We show that every quasi-alternating link gives rise to an infinite family of quasi-alternating links obtained by replacing a crossing with an alternating rational tangle. Consequently, we show that many pretzel links are quasi-alternating, and we determine the thickness of Khovanov homology for "most" pretzel links with arbitrarily many strands.

2004.08182

Stellar angular momentum distribution linked to galaxy morphology

We study the spatially-resolved stellar specific angular momentum $j_*$ in a high-quality sample of 24 CALIFA galaxies covering a broad range of visual morphology, accounting for stellar velocity and velocity dispersion. The shape of the spaxel-wise probability density function of normalised $s=j_*/j_{*mean}$, PDF($s$), deviates significantly from the near-universal initial distribution expected of baryons in a dark matter halo and can be explained by the expected baryonic effects in galaxy formation that remove and redistribute angular momentum. Further we find that the observed shape of the PDF($s$) correlates significantly with photometric morphology, where late-type galaxies have PDF($s$) that is similar to a normal distribution, whereas early types have a strongly-skewed PDF($s$) resulting from an excess of low-angular momentum material. Galaxies that are known to host pseudobulges (bulge Sersic index $n_b <2.2$) tend to have less skewed bulge PDF($s$), with skewness $(b_{1rb})\lesssim0.8$. The PDF($s$) encodes both kinematic and photometric information and appears to be a robust tracer of morphology. Its use is motivated by the desire to move away from traditional component-based classifications which are subject to observer bias, to classification on a galaxy's fundamental (stellar mass, angular momentum) properties. In future, PDF($s$) may also be useful as a kinematic decomposition tool.

Observations of young open clusters show a bimodal distribution of stellar rotation. Sun-like stars in those clusters group into two main sub-populations of fast and slow rotators. Beyond an age of about 500 Myrs, the two populations converge towards a single peak distribution of angular velocities. I argue that this evolution of stellar rotation in open clusters results from a brief episode of enhanced angular momentum loss by strong stellar wind during the early evolution of rapidly rotating Sun-like stars

2004.08183

Majority rule on rhombus tilings and Condorcet super-domains

In this paper we consider a Condorcet domain (CD) formed by a rhombus tiling as a voting design and consider a problem of aggregation of voting designs using majority rule. A Condorcet super-domain is a collection of CDs obtained from rhombus tilings on a zonogone Z(n; 2) with the property that if voting designs (ballots) belong to this collection, then the simple majority rule does not yield cycles. A study of Condorcet super-domains and methods of constructing them form the main subject of this paper.

We explore a method for computing admissible heuristic evaluation functions for search problems. It utilizes pattern databases, which are precomputed tables of the exact cost of solving various subproblems of an existing problem. Unlike standard pattern database heuristics, however, we partition our problems into disjoint subproblems, so that the costs of solving the different subproblems can be added together without overestimating the cost of solving the original problem. Previously, we showed how to statically partition the sliding-tile puzzles into disjoint groups of tiles to compute an admissible heuristic, using the same partition for each state and problem instance. Here we extend the method and show that it applies to other domains as well. We also present another method for additive heuristics which we call dynamically partitioned pattern databases. Here we partition the problem into disjoint subproblems for each state of the search dynamically. We discuss the pros and cons of each of these methods and apply both methods to three different problem domains: the sliding-tile puzzles, the 4-peg Towers of Hanoi problem, and finding an optimal vertex cover of a graph. We find that in some problem domains, static partitioning is most effective, while in others dynamic partitioning is a better choice. In each of these problem domains, either statically partitioned or dynamically partitioned pattern database heuristics are the best known heuristics for the problem.

2004.08184

FaVAD: A software workflow for characterisation and visualizing of defects in crystalline structures

The analysis of defects and defect dynamics in crystalline materials is important for fundamental science and for a wide range of applied engineering. With increasing system size the analysis of molecular-dynamics simulation data becomes non-trivial. Here, we present a workflow for semi-automatic identification and classification of defects in crystalline structures, combining a new approach for defect description with several already existing open-source software packages. Our approach addresses the key challenges posed by the often relatively tiny volume fraction of the modified parts of the sample, thermal motion and the presence of potentially unforeseen atomic configurations (defect types) after irradiation. The local environment of any atom is converted into a rotation-invariant descriptive vector ('fingerprint'), which can be compared to known defect types and also yields a distance metric suited for classification. Vectors which cannot be associated to known structures indicate new types of defects. As proof-of-concept we apply our method on an iron sample to analyze the defects caused by a collision cascade induced by a 10 keV primary-knock-on-atom. The obtained results are in good agreement with reported literature values.

Recent interest in exploring local vacuum structure of QCD through the properties of the eigenmodes of the lattice Dirac operators rises again the challenge to visualize four-dimensional objects and structures which appear in lattice field theories. In spite of complex and powerful commercial visualization software packages on the market, there are reasons to develop Interactive Visualization Package (IVP). We believe that an apprehension of the complex structures is possible only through the interactive approach, with the user being able to manipulate data representations and slices through the lattice in real-time. Further insight should also be gained by an interactive parallel examination of different physical quantities, e.g. eigenmode density with topological charge or action densities. Finally, thanks to constantly falling hardware prices, IVP makes it possible to use almost any Linux PC as a visualization tool for research in lattice field theory.

2004.08186

Thermodynamic equilibrium of massless fermions with vorticity, chirality and magnetic field

The present thesis aimed to examine the effects of vorticity on the thermodynamics of relativistic quantum systems. We extend the Zubarev's non-equilibrium statistical operator method to address quantum effects induced by vorticity in the presence of chiral matter and external electromagnetic field and keeping full covariant and quantum properties of the system. To investigate the effects of vorticity, this work has been focused on systems consisting of massless chiral fermions. We recovered the significant quantum phenomena known in the literature, namely the chiral magnetic effect, the chiral vortical effect, the axial vortical effect and the chiral separation effect and we also revealed the presence of additional effects at second-order on thermal vorticity. This study has also identified and presented the exact solutions of thermal states for a system at global thermal equilibrium consisting of chiral massless fermions under the action of an external constant homogeneous magnetic field. Taking advantage of these exact solutions and conservation equations, the study also proved that the thermal coefficients related to first-order effects on thermal vorticity do not receive corrections from the external electromagnetic field. The same argument revealed existing relations between those thermal coefficients, even connecting coefficients related to vorticity to other related to electromagnetic field. For instance, this analysis has found that the chiral vortical effect and the chiral magnetic effect conductivities are connected one to the other by a differential equation. Therefore, this research provides the first steps into deriving the relations between the effects and the interplay of electromagnetic fields and vorticity.

The chiral magnetic effect is a phenomenon where an electromagnetic current is generated along a magnetic field. Recently, in nonequilibrium systems, negative longitudinal magnetoresistance has been observed experimentally in Dirac/Weyl semimetals, which provides evidence for the chiral magnetic effect as a nonequilibrium current. On the other hand, the emergence of the chiral magnetic effect as an equilibrium current is still controversial. We propose a possible realization of the chiral magnetic effect as an equilibrium current using inhomogeneous magnetic fields. By employing tight-binding calculations and linear response theory, we demonstrate that a finite current density is generated by inhomogeneous magnetic fields, while the spatial integration of the current is equal to zero, which is consistent with the so-called "no-go theorem" of the chiral magnetic effect in real lattice systems. Moreover, we propose an experimental setup to detect the effect in Weyl semimetal materials.

2004.08187

Graphical complexes of groups

We introduce graphical complexes of groups, which can be thought of as a generalisation of Coxeter systems with 1-dimensional nerves. We show that these complexes are strictly developable, and we equip the resulting Basic Construction with three structures of non-positive curvature: piecewise linear CAT(0), C(6) graphical small cancellation, and a systolic one. We then use these structures to establish various properties of the fundamental groups of these complexes, such as biautomaticity and Tits Alternative. We isolate an easily checkable condition implying hyperbolicity of the fundamental groups, and we construct some non-hyperbolic examples. We also briefly discuss a parallel theory of C(4)-T(4) graphical complexes of groups and outline their basic properties.

Undirected graphical models, or Markov networks, are a popular class of statistical models, used in a wide variety of applications. Popular instances of this class include Gaussian graphical models and Ising models. In many settings, however, it might not be clear which subclass of graphical models to use, particularly for non-Gaussian and non-categorical data. In this paper, we consider a general sub-class of graphical models where the node-wise conditional distributions arise from exponential families. This allows us to derive multivariate graphical model distributions from univariate exponential family distributions, such as the Poisson, negative binomial, and exponential distributions. Our key contributions include a class of M-estimators to fit these graphical model distributions; and rigorous statistical analysis showing that these M-estimators recover the true graphical model structure exactly, with high probability. We provide examples of genomic and proteomic networks learned via instances of our class of graphical models derived from Poisson and exponential distributions.

2004.08188

Ramsey-biased spectroscopy of superconducting qubits under dispersion

We proposed a spectroscopic method that extends Ramsey's atomic spectroscopy to detect the transition frequency of a qubit fabricated on a superconducting circuit. The method uses a multi-interval train of qubit biases to implement an alternate resonant and dispersive couplings to an incident probe field. The consequent absorption spectrum of the qubit has a narrower linewidth at its transition frequency than that obtained from constantly biasing the qubit to resonance while the middle dispersive evolution incurs only a negligible shift in detected frequency. Modeling on transmon qubits, we find that the linewidth reduction reaches 23% and Ramsey fringes are simultaneously suppressed at extreme duration ratio of dispersion over resonance for a double-resonance scheme. If the scheme is augmented by an extra resonance segment, a further 37% reduction can be achieved.

We introduce a model of quantum magnetism induced by the non-perturbative exchange of microwave photons between distant superconducting qubits. By interconnecting qubits and cavities, we obtain a spin-boson lattice model that exhibits a quantum phase transition where both qubits and cavities spontaneously polarise. We present a many-body ansatz that captures this phenomenon all the way, from a the perturbative dispersive regime where photons can be traced out, to the non-perturbative ultra-strong coupling regime where photons must be treated on the same footing as qubits. Our ansatz also reproduces the low-energy excitations, which are described by hybridised spin-photon quasiparticles, and can be probed spectroscopically from transmission experiments in circuit-QED, as shown by simulating a possible experiment by Matrix-Product-State methods.

2004.08189

MOPT: Multi-Object Panoptic Tracking

Comprehensive understanding of dynamic scenes is a critical prerequisite for intelligent robots to autonomously operate in our environment. Research in this domain which encompasses diverse perception problems has primarily been limited to addressing specific tasks individually and thus has contributed very little towards modeling the ability to understand dynamic scenes holistically. As a step towards encouraging research in this direction, we introduce a new perception task that we name Multi-Object Panoptic Tracking (MOPT). MOPT unifies the conventionally disjoint tasks of semantic segmentation, instance segmentation, and multi-object tracking. MOPT allows for exploiting pixel-level semantic information of 'thing' and 'stuff' classes, temporal coherence, and pixel-level associations over time, for the mutual benefit of each of these sub-problems. In order to facilitate quantitative evaluations of MOPT in a unified manner, we propose the soft Panoptic Tracking Quality (sPTQ) metric. As a first step towards addressing this task, we propose the novel PanopticTrackNet architecture that builds upon the state-of-the-art top-down panoptic segmentation network EfficientPS by adding a new tracking head to simultaneously learn all subtasks in an end-to-end manner. Additionally, we present several strong baselines that combine predictions from state-of-the-art panoptic segmentation and multi-object tracking models for comparison. We present extensive quantitative and qualitative evaluations for both vision-based and LiDAR-based MOPT on the challenging Virtual KITTI 2 and SemanticKITTI datasets, which demonstrates encouraging results.

We describe the recently introduced extremal optimization algorithm and apply it to target detection and association problems arising in pre-processing for multi-target tracking. Here we consider the problem of pre-processing for multiple target tracking when the number of sensor reports received is very large and arrives in large bursts. In this case, it is sometimes necessary to pre-process reports before sending them to tracking modules in the fusion system. The pre-processing step associates reports to known tracks (or initializes new tracks for reports on objects that have not been seen before). It could also be used as a pre-process step before clustering, e.g., in order to test how many clusters to use. The pre-processing is done by solving an approximate version of the original problem. In this approximation, not all pair-wise conflicts are calculated. The approximation relies on knowing how many such pair-wise conflicts that are necessary to compute. To determine this, results on phase-transitions occurring when coloring (or clustering) large random instances of a particular graph ensemble are used.

2004.08190

Structured Landmark Detection via Topology-Adapting Deep Graph Learning

Image landmark detection aims to automatically identify the locations of predefined fiducial points. Despite recent success in this filed, higher-ordered structural modeling to capture implicit or explicit relationships among anatomical landmarks has not been adequately exploited. In this work, we present a new topology-adapting deep graph learning approach for accurate anatomical facial and medical (e.g., hand, pelvis) landmark detection. The proposed method constructs graph signals leveraging both local image features and global shape features. The adaptive graph topology naturally explores and lands on task-specific structures which is learned end-to-end with two Graph Convolutional Networks (GCNs). Extensive experiments are conducted on three public facial image datasets (WFLW, 300W and COFW-68) as well as three real-world X-ray medical datasets (Cephalometric (public), Hand and Pelvis). Quantitative results comparing with the previous state-of-the-art approaches across all studied datasets indicating the superior performance in both robustness and accuracy. Qualitative visualizations of the learned graph topologies demonstrate a physically plausible connectivity laying behind the landmarks.

Getting deep convolutional neural networks to perform well requires a large amount of training data. When the available labelled data is small, it is often beneficial to use transfer learning to leverage a related larger dataset (source) in order to improve the performance on the small dataset (target). Among the transfer learning approaches, domain adaptation methods assume that distributions between the two domains are shifted and attempt to realign them. In this paper, we consider the domain adaptation problem from the perspective of dimensionality reduction and propose a generic framework based on graph embedding. Instead of solving the generalised eigenvalue problem, we formulate the graph-preserving criterion as a loss in the neural network and learn a domain-invariant feature transformation in an end-to-end fashion. We show that the proposed approach leads to a powerful Domain Adaptation framework; a simple LDA-inspired instantiation of the framework leads to state-of-the-art performance on two of the most widely used Domain Adaptation benchmarks, Office31 and MNIST to USPS datasets.

2004.08191

Adjoint algebraic groups as automorphism groups of a projector on a central simple algebra

We show that any adjoint absolutely simple linear algebraic group over a field of characteristic zero is the automorphism group of some projector on a central simple algebra. Projective homogeneous varieties can be described in these terms; in particular, we reproduce quadratic equations by Lichtenstein defining them.

We introduce a general class of automorphisms of rotation algebras, the noncommutative Furstenberg transformations. We prove that fully irrational noncommutative Furstenberg transformations have the tracial Rokhlin property, which is a strong form of outerness. We conclude that crossed products by these automorphisms have stable rank one, real rank zero, and order on projections determined by traces (Blackadar's Second Fundamental Comparability Question). We also prove that several classes of simple quotients of the C*-algebras of discrete subgroups of five dimensional nilpotent Lie groups, considered by Milnes and Walters, are crossed products of simple C*-algebras (C*-algebras of minimal ordinary Furstenberg transformations) by automorphisms which have the tracial Rokhlin property. It follows that these algebras also have stable rank one, real rank zero, and order on projections determined by traces.

2004.08192

The Unusual Properties of Water Confined in Soft and Heterogeneous Nanopores

Liquids under confinement differ in behavior from their bulk counterparts and can acquire properties that are specific to the confined phase and linked to the nature and structure of the host matrix. While confined liquid water is not a new topic of research, the past few years have seen a series of intriguing novel features for water inside nanoscale pores. These unusual properties arise from the very specific nature of nanoporous materials: termed "soft porous crystals", they combine large-scale flexibility with a heterogeneous internal surface. This creates a rich diversity of behavior for the adsorbed water, and the combination of different experimental characterization techniques along with computational chemistry at various scales is necessary to understand the phenomena observed, and their microscopic origins. The range of systems of interest span the whole chemical range, from the inorganic (zeolites, imogolites) to the organic (microporous carbons, graphene and its derivatives) and even encompass the hybrid organic--inorganic systems (such as metal--organic frameworks).

We present experimental evidences for the existence of a semi-hard scale in light hadrons. This includes the suppression of gluon radiation that is seen in high mass hadron diffraction; the weak energy dependence of hadronic total cross sections; the small value of the Pomeron trajectory slope measured in photoproduction of J/Psi; the weakness of gluon shadowing in nuclei; shortage of gluons in the proton revealed by an unusual behavior of the proton structure function in the soft limit, and the enhanced intrinsic transverse momentum of quarks and gluons, which considerably exceeds the inverse hadronic size. All these observations suggest that gluons in hadrons are located within spots of a small size relative to the confinement radius.

2004.08193

An Annotated Dataset of Stack Overflow Post Edits

To improve software engineering, software repositories have been mined for code snippets and bug fixes. Typically, this mining takes place at the level of files or commits. To be able to dig deeper and to extract insights at a higher resolution, we hereby present an annotated dataset that contains over 7 million edits of code and text on Stack Overflow. Our preliminary study indicates that these edits might be a treasure trove for mining information about fine-grained patches, e.g., for the optimisation of non-functional properties.

Question-answer driven Semantic Role Labeling (QA-SRL) has been proposed as an attractive open and natural form of SRL, easily crowdsourceable for new corpora. Recently, a large-scale QA-SRL corpus and a trained parser were released, accompanied by a densely annotated dataset for evaluation. Trying to replicate the QA-SRL annotation and evaluation scheme for new texts, we observed that the resulting annotations were lacking in quality and coverage, particularly insufficient for creating gold standards for evaluation. In this paper, we present an improved QA-SRL annotation protocol, involving crowd-worker selection and training, followed by data consolidation. Applying this process, we release a new gold evaluation dataset for QA-SRL, yielding more consistent annotations and greater coverage. We believe that our new annotation protocol and gold standard will facilitate future replicable research of natural semantic annotations.

2004.08194

Learning-Based Joint User-AP Association and Resource Allocation in Ultra Dense Network

With the advantages of Millimeter wave in wireless communication network, the coverage radius and inter-site distance can be further reduced, the ultra dense network (UDN) becomes the mainstream of future networks. The main challenge faced by UDN is the serious inter-site interference, which needs to be carefully addressed by joint user association and resource allocation methods. In this paper, we propose a multi-agent Q-learning based method to jointly optimize the user association and resource allocation in UDN. The deep Q-network is applied to guarantee the convergence of the proposed method. Simulation results reveal the effectiveness of the proposed method and different performances under different simulation parameters are evaluated.

With the fast development of Internet of things (IoT), the fifth generation (5G) wireless networks need to provide massive connectivity of IoT devices and meet the demand for low latency. To satisfy these requirements, Non-Orthogonal Multiple Access (NOMA) has been recognized as a promising solution for 5G networks to significantly improve the network capacity. In parallel with the development of NOMA techniques, Mobile Edge Computing (MEC) is becoming one of the key emerging technologies to reduce the latency and improve the Quality of Service (QoS) for 5G networks. In order to capture the potential gains of NOMA in the context of MEC, this paper proposes an edge computing aware NOMA technique which can enjoy the benefits of uplink NOMA in reducing MEC users' uplink energy consumption. To this end, we formulate a NOMA based optimization framework which minimizes the energy consumption of MEC users via optimizing the user clustering, computing and communication resource allocation, and transmit powers. In particular, similar to frequency Resource Blocks (RBs), we divide the computing capacity available at the cloudlet to computing RBs. Accordingly, we explore the joint allocation of the frequency and computing RBs to the users that are assigned to different order indices within the NOMA clusters. We also design an efficient heuristic algorithm for user clustering and RBs allocation, and formulate a convex optimization problem for the power control to be solved independently per NOMA cluster. The performance of the proposed NOMA scheme is evaluated via simulations.

2004.08195

The FaceChannel: A Light-weight Deep Neural Network for Facial Expression Recognition

Current state-of-the-art models for automatic FER are based on very deep neural networks that are difficult to train. This makes it challenging to adapt these models to changing conditions, a requirement from FER models given the subjective nature of affect perception and understanding. In this paper, we address this problem by formalizing the FaceChannel, a light-weight neural network that has much fewer parameters than common deep neural networks. We perform a series of experiments on different benchmark datasets to demonstrate how the FaceChannel achieves a comparable, if not better, performance, as compared to the current state-of-the-art in FER.

We present a new action recognition deep neural network which adaptively learns the best action velocities in addition to the classification. While deep neural networks have reached maturity for image understanding tasks, we are still exploring network topologies and features to handle the richer environment of video clips. Here, we tackle the problem of multiple velocities in action recognition, and provide state-of-the-art results for gesture recognition, on known and new collected datasets. We further provide the training steps for our semi-supervised network, suited to learn from huge unlabeled datasets with only a fraction of labeled examples.

2004.08196

Systems with size and energy polydispersity: from glasses to mosaic crystals

We use Langevin dynamics simulations to study dense 2d systems of particles with both size and energy polydispersity. We compare two types of bidisperse systems which differ in the correlation between particle size and interaction parameters: in one system big particles have high interaction parameters and small particles have low interaction parameters, while in the other system the situation is reversed. We study the different phases of the two systems and compare them to those of a system with size but not energy bidispersity. We show that, depending on the strength of interaction between big and small particles, cooling to low temperatures yields either homogeneous glasses or mosaic crystals.

Transparent glasses of BaNaB9O15 (BNBO) were fabricated via the conventional melt-quenching technique. The amorphous and the glassy nature of the as-quenched samples were respectively, confirmed by X-ray powder diffraction (XRD) and differential scanning calorimetry (DSC). The glass transition and crystallization parameters were evaluated under non-isothermal conditions using DSC. The correlation between the heating rate dependent glass transition and the crystallization temperatures was discussed and deduced the Kauzmann temperature for BNBO glass-plates and powdered samples. The values of the Kauzmann temperature for the plates and powdered samples were 776 K and 768 K, respectively. Approximation-free method was used to evaluate the crystallization kinetic parameters for the BNBO glass samples. The effect of the sample thickness on the crystallization kinetics of BNBO glasses was also investigated.

2004.08197

Long-time asymptotic behaviour of the value function in nonlinear stopping problems

We provide general conditions ensuring that the value functions of some nonlinear stopping problems with finite horizon converge to the value functions of the corresponding problems with infinite horizon. We also provide the rate of convergence. Our results can be formulated as results on stability, with respect to time horizon, of nonlinear $f$-expectations. Applications to stopping problems involving some integro-differential operators are given. They include asymptotic behaviour of fair prices of American options in an exponential L\'evy model.

In this paper, we study the Cauchy problem with decaying initial data for the nonlocal modified Korteweg-de Vries equation (nonlocal mKdV) \[q_t(x,t)+q_{xxx}(x,t)-6q(x,t)q(-x,-t)q_x(x,t)=0,\] which can be viewed as a generalization of the local classical mKdV equation. We first formulate the Riemann-Hilbert problem associated with the Cauchy problem of the nonlocal mKdV equation. Then we apply the Deift-Zhou nonlinear steepest-descent method to analyze the long-time asymptotics for the solution of the nonlocal mKdV equation. In contrast with the classical mKdV equation, we find some new and different results on long-time asymptotics for the nonlocal mKdV equation and some additional assumptions about the scattering data are made in our main results.

2004.08199

On the equivariant $K$- and $KO$-homology of some special linear groups

We compute the equivariant $KO$-homology of the classifying space for proper actions of $\textrm{SL}_3(\mathbb{Z})$ and $\textrm{GL}_3(\mathbb{Z})$. We also compute the Bredon homology and equivariant $K$-homology of the classifying spaces for proper actions of $\textrm{PSL}_2(\mathbb{Z}[\frac{1}{p}])$ and $\textrm{SL}_2(\mathbb{Z}[\frac{1}{p}])$ for each prime $p$. Finally, we prove the unstable Gromov-Lawson-Rosenberg conjecture for $\textrm{PSL}_2(\mathbb{Z}[\frac{1}{p}])$ when $p\equiv11\pmod{12}$.

For a rational homology 3-sphere $Y$ with a $\spinc$ structure $\s$, we show that simple algebraic manipulations of our construction of equivariant Seiberg-Witten Floer homology lead to a collection of variants which are topological invariants. We establish exact sequences relating them, we show that they satisfy a duality under orientation reversal, and we explain their relation to ou previous construction of equivariant Seiberg-Witten Floer (co)homologies. We conjecture the equivalence of these versions of equivariant Seiberg-Witten Floer homology with the Heegaard Floer invariants introduced by Ozsv\'ath and Szab\'o.

2004.08200

Defining and Verifying Durable Opacity: Correctness for Persistent Software Transactional Memory

Non-volatile memory (NVM), aka persistent memory, is a new paradigm for memory that preserves its contents even after power loss. The expected ubiquity of NVM has stimulated interest in the design of novel concepts ensuring correctness of concurrent programming abstractions in the face of persistency. So far, this has lead to the design of a number of persistent concurrent data structures, built to satisfy an associated notion of correctness: durable linearizability. In this paper, we transfer the principle of durable concurrent correctness to the area of software transactional memory (STM). Software transactional memory algorithms allow for concurrent access to shared state. Like linearizability for concurrent data structures, opacity is the established notion of correctness for STMs. First, we provide a novel definition of durable opacity extending opacity to handle crashes and recovery in the context of NVM. Second, we develop a durably opaque version of an existing STM algorithm, namely the Transactional Mutex Lock (TML). Third, we design a proof technique for durable opacity based on refinement between TML and an operational characterisation of durable opacity by adapting the TMS2 specification. Finally, we apply this proof technique to show that the durable version of TML is indeed durably opaque. The correctness proof is mechanized within Isabelle.

Data hosted in a cloud environment can be subject to attacks from a higher privileged adversary, such as a malicious or compromised cloud provider. To provide confidentiality and integrity even in the presence of such an adversary, a number of Trusted Execution Environments (TEEs) have been developed. A TEE aims to protect data and code within its environment against high privileged adversaries, such as a malicious operating system or hypervisor. While mechanisms exist to attest a TEE's integrity at load time, there are no mechanisms to attest its integrity at runtime. Additionally, work also exists that discusses mechanisms to verify the runtime integrity of programs and system components. However, those verification mechanisms are themselves not protected against attacks from a high privileged adversary. It is therefore desirable to combine the protection mechanisms of TEEs with the ability of application runtime integrity verification. In this paper, we present Scanclave, a lightweight design which achieves three design goals: Trustworthiness of the verifier, a minimal trusted software stack and the possibility to access an application's memory from a TEE. Having achieved our goals, we are able to verify the runtime integrity of applications even in the presence of a high privileged adversary. We refrain from discussing which properties define the runtime integrity of an application, as different applications will require different verification methods. Instead, we show how Scanclave enables a remote verifier to determine the runtime integrity of an application. Afterwards, we perform a security analysis for the different steps of our design. Additionally, we discuss different enclave implementations that might be used for the implementation of Scanclave.

2004.08201

ERC20 Transactions over Ethereum Blockchain: Network Analysis and Predictions

Following the birth of Bitcoin and the introduction of the Ethereum ERC20 protocol a decade ago, recent years have witnessed a growing number of cryptographic tokens that are being introduced by researchers, private sector companies and NGOs. The ubiquitous of such Blockchain based cryptocurrencies give birth to a new kind of rising economy, which presents great difficulties to modeling its dynamics using conventional semantic properties. Our work presents the analysis of the dynamical properties of the ERC20 protocol compliant crypto-coins' trading data using a network theory prism. We examine the dynamics of ERC20 based networks over time by analyzing a meta-parameter of the network, the power of its degree distribution. Our analysis demonstrates that this parameter can be modeled as an under-damped harmonic oscillator over time, enabling a year forward of network parameters predictions.

It is undeniable that artificial intelligence (AI) and blockchain concepts are spreading at a phenomenal rate. Both technologies have distinct degree of technological complexity and multi-dimensional business implications. However, a common misunderstanding about blockchain concept, in particular, is that blockchain is decentralized and is not controlled by anyone. But the underlying development of a blockchain system is still attributed to a cluster of core developers. Take smart contract as an example, it is essentially a collection of codes (or functions) and data (or states) that are programmed and deployed on a blockchain (say, Ethereum) by different human programmers. It is thus, unfortunately, less likely to be free of loopholes and flaws. In this article, through a brief overview about how artificial intelligence could be used to deliver bug-free smart contract so as to achieve the goal of blockchain 2.0, we to emphasize that the blockchain implementation can be assisted or enhanced via various AI techniques. The alliance of AI and blockchain is expected to create numerous possibilities.

2004.08202

Women worry about family, men about the economy: Gender differences in emotional responses to COVID-19

Among the critical challenges around the COVID-19 pandemic is dealing with potentially detrimental effects on people's mental health. Designing appropriate interventions and identifying the concerns of those most at risk requires methods that can extract worries, concerns and emotional responses from text data. We examine gender differences and the effect of document length on worries about the ongoing COVID-19 situation. Our findings suggest that i) shorter texts do not offer an as adequate insight into psychological processes as longer texts. We further find ii) marked gender differences in topics concerning emotional responses. Women worried more about their loved ones and severe health concerns while men were more occupied with effects on the economy and society. The findings align with general gender differences in language found elsewhere, but the current unique circumstances likely amplified these effects. We close this paper with a call for more high-quality datasets due to the limitations of Tweet-sized data.

The paper studies different regression approaches for modeling COVID-19 spread and its impact on the stock market. The logistic curve model was used with Bayesian regression for predictive analytics of the coronavirus spread. The impact of COVID-19 was studied using regression approach and compared to other crises influence. In practical analytics, it is important to find the maximum of coronavirus cases per day, this point means the estimated half time of coronavirus spread in the region under investigation. The obtained results show that different crises with different reasons have different impact on the same stocks. It is important to analyze their impact separately. Bayesian inference makes it possible to analyze the uncertainty of crisis impacts.

2004.08203

'Bonn' Optimized Stellar Tracks (BoOST). Simulated Populations of Massive and Very Massive Stars as Input for Astrophysical Applications

Massive and very massive stars can play important roles in stellar populations by ejecting strong stellar winds and exploding in energetic phenomena. It is therefore of high importance that their behaviour is properly accounted for in synthetic model populations. Here we present nine grids of stellar evolutionary model sequences, together with finely resolved interpolated sequences and synthetic populations, of stars with 9-500 Msun and with metallicities ranging from Solar down to 1/250 Solar. The stellar models were computed with the 'Bonn' evolutionary code (covering core-hydrogen- and core-helium-burning phases, both complete). Post-processing for publication has been done using optimized methods, developed by our team, for massive and very massive stars. Interpolation and population synthesis were also performed on the models by our newly developed routine synStars. Eight of the grids represent slowly rotating massive stars with normal/classical evolution, while one grid represents fast rotating, chemically-homogeneously evolving models. Apart from the common stellar parameters such as mass, radius, surface temperature, luminosity and mass loss rate, we present stellar wind properties such as estimated wind velocity and kinetic energy of the wind. Additionally, we provide complete chemical yields of 34 isotopes, and estimates for the masses of the compact object remnants. The 'Bonn' Optimized Stellar Tracks (BoOST) project is published as simple tables - including stellar models, interpolated tracks and synthetic populations - thus ideal for further scientific applications. For example, star-formation studies could be done with BoOST to cover broad metallicity ranges, and so could be simulations of high-redshift galaxies. Additionally, gravitational-wave event rate predictions could be refined using BoOST by accounting for very massive stars at low-metallicity.

Very massive primordial stars (140 Msol < M < 260 Msol) are supposed to end their lives as PISN. Such an event can be traced by a typical chemical signature in low metallicity stars, but at the present time, this signature is lacking in the extremely metal-poor stars we are able to observe. Does it mean that those very massive objects were not formed, contrarily to the primordial star formation scenarios ? Could it be possible that they avoided this tragical fate ? We explore the effects of rotation, anisotropical mass loss and magnetic field on the core size of very massive Population III models. We find that magnetic fields provide the strong coupling that is lacking in standard evolution metal-free models and our 150 Msol Population III model avoids indeed the pair-instability explosion.

2004.08204

Modeling Institutional Credit Risk with Financial News

Credit risk management, the practice of mitigating losses by understanding the adequacy of a borrower's capital and loan loss reserves, has long been imperative to any financial institution's long-term sustainability and growth. MassMutual is no exception. The company is keen on effectively monitoring downgrade risk, or the risk associated with the event when credit rating of a company deteriorates. Current work in downgrade risk modeling depends on multiple variations of quantitative measures provided by third-party rating agencies and risk management consultancy companies. As these structured numerical data become increasingly commoditized among institutional investors, there has been a wide push into using alternative sources of data, such as financial news, earnings call transcripts, or social media content, to possibly gain a competitive edge in the industry. The volume of qualitative information or unstructured text data has exploded in the past decades and is now available for due diligence to supplement quantitative measures of credit risk. This paper proposes a predictive downgrade model using solely news data represented by neural network embeddings. The model standalone achieves an Area Under the Receiver Operating Characteristic Curve (AUC) of more than 80 percent. The output probability from this news model, as an additional feature, improves the performance of our benchmark model using only quantitative measures by more than 5 percent in terms of both AUC and recall rate. A qualitative evaluation also indicates that news articles related to our predicted downgrade events are specially relevant and high-quality in our business context.

The financial services industry has unique explainability and fairness challenges arising from compliance and ethical considerations in credit decisioning. These challenges complicate the use of model machine learning and artificial intelligence methods in business decision processes.

2004.08205

Large-scale analysis of grooming in modern social networks

Social networks are evolving to engage their users more by providing them with more functionalities. One of the most attracting ones is streaming. Users may broadcast part of their daily lives to thousands of others world-wide and interact with them in real-time. Unfortunately, this feature is reportedly exploited for grooming. In this work, we provide the first in-depth analysis of this problem for social live streaming services. More precisely, using a dataset that we collected, we identify predatory behaviours and grooming on chats that bypassed the moderation mechanisms of the LiveMe, the service under investigation. Beyond the traditional text approaches, we also investigate the relevance of emojis in this context, as well as the user interactions through the gift mechanisms of LiveMe. Finally, our analysis indicates the possibility of grooming towards minors, showing the extent of the problem in such platforms.

Social Networking Site (SNS) is a great innovation of modern times. Facebook, Twitter etc. have become an everyday part of peoples' life. Among all SNSs, Facebook is the most popular social network all over the world. Bangladesh is no exception. People of Bangladesh use Facebook for social communication, online shopping, business, knowledge and experience sharing etc. As well as the various uses of SNSs, people sometimes find themselves involved in real life violence, provoked by some social media posts or activities. In this paper, we discussed some case studies in which real life violence is originated based on Facebook activities in Bangladesh. Facebook was used in these incidents intentionally or unintentionally mostly as a tool to trigger hatred and violence. We analyzed and discussed the real-world consequences of these virtual activities in social media. Lastly, we recommended possible future measurements to prevent such violence.

2004.08206

Vehicle Position Estimation with Aerial Imagery from Unmanned Aerial Vehicles

The availability of real-world data is a key element for novel developments in the fields of automotive and traffic research. Aerial imagery has the major advantage of recording multiple objects simultaneously and overcomes limitations such as occlusions. However, there are only few data sets available. This work describes a process to estimate a precise vehicle position from aerial imagery. A robust object detection is crucial for reliable results, hence the state-of-the-art deep neural network Mask-RCNN is applied for that purpose. Two training data sets are employed: The first one is optimized for detecting the test vehicle, while the second one consists of randomly selected images recorded on public roads. To reduce errors, several aspects are accounted for, such as the drone movement and the perspective projection from a photograph. The estimated position is comapared with a reference system installed in the test vehicle. It is shown, that a mean accuracy of 20 cm can be achieved with flight altitudes up to 100 m, Full-HD resolution and a frame-by-frame detection. A reliable position estimation is the basis for further data processing, such as obtaining additional vehicle state variables. The source code, training weights, labeled data and example videos are made publicly available. This supports researchers to create new traffic data sets with specific local conditions.

This letter studies a wireless system consisting of distributed ground terminals (GTs) communicating with an unmanned aerial vehicle (UAV) that serves as a mobile base station (BS). The UAV flies cyclically above the GTs at a fixed altitude, which results in a cyclical pattern of the strength of the UAV-GT channels. To exploit such periodic channel variations, we propose a new cyclical multiple access (CMA) scheme to schedule the communications between the UAV and GTs in a cyclical time-division manner based on the flying UAV's position. The time allocations to different GTs are optimized to maximize their minimum throughput. It is revealed that there is a fundamental tradeoff between throughput and access delay in the proposed CMA. Simulation results show significant throughput gains over the case of a static UAV BS in delay-tolerant applications.

2004.08207

Simulation of Covid-19 epidemic evolution: are compartmental models really predictive?

Computational models for the simulation of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemic evolution would be extremely useful to support authorities in designing healthcare policies and lockdown measures to contain its impact on public health and economy. In Italy, the devised forecasts have been mostly based on a pure data-driven approach, by fitting and extrapolating open data on the epidemic evolution collected by the Italian Civil Protection Center. In this respect, SIR epidemiological models, which start from the description of the nonlinear interactions between population compartments, would be a much more desirable approach to understand and predict the collective emergent response. The present contribution addresses the fundamental question whether a SIR epidemiological model, suitably enriched with asymptomatic and dead individual compartments, could be able to provide reliable predictions on the epidemic evolution. To this aim, a machine learning approach based on particle swarm optimization (PSO) is proposed to automatically identify the model parameters based on a training set of data of progressive increasing size, considering Lombardy in Italy as a case study. The analysis of the scatter in the forecasts shows that model predictions are quite sensitive to the size of the dataset used for training, and that further data are still required to achieve convergent -- and therefore reliable -- predictions.

With the world-wide development of 2019 novel coronavirus, although WHO has officially announced the disease as COVID-19, one controversial term - "Chinese Virus" is still being used by a great number of people. In the meantime, global online media coverage about COVID-19-related racial attacks increases steadily, most of which are anti-Chinese or anti-Asian. As this pandemic becomes increasingly severe, more people start to talk about it on social media platforms such as Twitter. When they refer to COVID-19, there are mainly two ways: using controversial terms like "Chinese Virus" or "Wuhan Virus", or using non-controversial terms like "Coronavirus". In this study, we attempt to characterize the Twitter users who use controversial terms and those who use non-controversial terms. We use the Tweepy API to retrieve 17 million related tweets and the information of their authors. We find significant differences between these two groups of Twitter users across their demographics, user-level features like the number of followers, political following status, as well as their geo-locations. Moreover, we apply classification models to predict Twitter users who are more likely to use controversial terms. To our best knowledge, this is the first large-scale social media-based study to characterize users with respect to their usage of controversial terms during a major crisis.

2004.08208

Contrast modes in a 3D ion transmission approach at keV energies

We present options for visualizing contrast maps in 3D ion transmission experiments. Simultaneous measurement of angular distributions and flight time of ions transmitted through self-supporting, single-crystalline silicon foils allows for mapping of intensity and different energy loss moments. The transmitted projectiles were detected mainly for random beam-sample orientation using pulsed beams of He ions and protons with incident energies 50 and 200 keV. Differences in contrast, observed when varying the projectile type and energy, can be attributed to sample nuclear and electronic structure and bear witness to impact parameter dependent energy loss processes. Our results provide a base for interpretation of data obtained in prospective transmission studies when for example using a helium ion microscope.

Scalability and performance of current flash memories can be improved substantially by replacing the floating poly-Si gate by a layer of Si dots. This multi-dot layer can be fabricated CMOS-compatibly in very thin gate oxide by ion beam synthesis (IBS). Here, we present both experimental and theoretical studies on IBS of multi-dot layers consisting of Si nanocrystals (NCs). The NCs are produced by ultra low energy Si ion implantation, which causes a high Si supersaturation in the shallow implantation region. During post-implantation annealing, this supersaturation leads to phase separation of the excess Si from the SiO2. Till now, the study of this phase separation process suffered from the weak Z contrast between Si and SiO2 in Transmission Electron Microscopy (TEM). Here, this imaging problem is resolved by mapping Si plasmon losses with a Scanning Transmission Electron Microscopy equipped with a parallel Electron Energy Loss Spectroscopy system (PEELS-STEM). Additionally, kinetic lattice Monte Carlo simulations of Si phase separation have been performed and compared with the experimental Si plasmon maps. It has been predicted theoretically that the morphology of the multi-dot Si floating-gate changes with increasing ion fluence from isolated, spherical NCs to percolated spinodal Si pattern. These patterns agree remarkably with PEELS-STEM images. However, the predicted fluence for spinodal patterns is lower than the experimental one. Because oxidants of the ambient atmosphere penetrate into the as-implanted SiO2, a substantial fraction of the implanted Si might be lost due to oxidation.

2004.08209

Dimension-and lattice-diversified Coulomb excitations in3D,2D,1D-nanotube electron gases,graphene and carbon nanotube

Elementary electronic excitations, which are due to the Coulomb-field scatterings, present the diverse phenomena in 3D, 2D, 1D-nanotube electron gases, graphene and carbon nanotubes. The critical mechanisms cover the dimension-dependent bare Coulomb potentials, energy dispersions, and free/valence carrier density. They are responsible for the main features, the available excitation channels (the electron-hole regions), the joint van Hove singularities, the undamped/damped collective excitations at small/sufficiently high transferred momenta, the momentum dependences of plasmon frequencies (acoustic and optical modes), and their categories (the intraband and inter-pi-band plasmons). There exists certain significant similarities and difference among various systems. The (momentum/ angular momentum, frequency)-excitation phase diagrams are directly reflected in the propagation of plasma waves.

An analytical method to investigate wave propagation in single- and double- walled carbon nanotubes under initial compression stress is presented. The nanotube structures are treated within the multilayer thin shell approximation with the elastic properties taken to be those of the graphene sheet. The governing equations are derived based on Flugge equations of motion. Frequency equations of wave propagation in single and double wall carbon nanotubes are described through the effects of initial compression stress and van der Waals force. To show the effects of Initial compression stress on the wave propagation in nanotubes, the symmetrical mode can be analyzed based on the present elastic continuum model. It is shown that the wave speed are sensitive to the compression stress especially for the lower frequencies.

2004.08210

A Class of Models with the Potential to Represent Fundamental Physics

A class of models intended to be as minimal and structureless as possible is introduced. Even in cases with simple rules, rich and complex behavior is found to emerge, and striking correspondences to some important core known features of fundamental physics are seen, suggesting the possibility that the models may provide a new approach to finding a fundamental theory of physics.

We develop a constructive approach to generate artificial neural networks representing the exact ground states of a large class of many-body lattice Hamiltonians. It is based on the deep Boltzmann machine architecture, in which two layers of hidden neurons mediate quantum correlations among physical degrees of freedom in the visible layer. The approach reproduces the exact imaginary-time Hamiltonian evolution, and is completely deterministic. In turn, compact and exact network representations for the ground states are obtained without stochastic optimization of the network parameters. The number of neurons grows linearly with the system size and total imaginary time, respectively. Physical quantities can be measured by sampling configurations of both physical and neuron degrees of freedom. We provide specific examples for the transverse-field Ising and Heisenberg models by implementing efficient sampling. As a compact, classical representation for many-body quantum systems, our approach is an alternative to the standard path integral, and it is potentially useful also to systematically improve on numerical approaches based on the restricted Boltzmann machine architecture.

2004.08211

Accelerating Universe in Hybrid and Logarithmic Teleparallel Gravity

Teleparallel gravity is a modified theory of gravity for which the Ricci scalar $R$ of the underlying geometry in the action is replaced by an arbitrary functional form of torsion scalar $T$. In doing so, cosmology in $% f(T)$ gravity becomes greatly simplified owing to the fact that $T$ contains only the first derivatives of the vierbeins. The article exploits this appealing nature of $f(T)$ gravity and present cosmological scenarios from hybrid and logarithmic teleparallel gravity models of the form $% f=e^{mT}T^n $ and $f=D\log(bT)$ respectively, where $m$, $n$, $D$ and $b$ are free parameters constrained to suffice the late time acceleration. We employ a well motivated parametrization of the deceleration parameter having just one degree of freedom constrained with a $\chi^{2}$ test from 57 data points of Hubble data set in the redshift range $0.07<z<2.36$, to obtain the expressions of pressure, density and EoS parameter for both the teleparallel gravity models and study their temporal evolution. We find the deceleration parameter to experience a signature flipping for the $\chi^{2}$ value of the free parameter at $z_{tr}\simeq0.6$ which is consistent with latest Planck measurements. Next, we present few geometric diagnostics of this parametrization to understand the nature of dark energy and its deviation from the $\Lambda$CDM cosmology. Finally, we study the energy conditions to check the consistency of the parameter spaces for both the teleparallel gravity models. We find the SEC to violate for both the models which is an essential recipe to obtain an accelerating universe.

We perform a phase space analysis of a generalized modified gravity theory with nonminimally coupling between geometry and matter. We apply the dynamical system approach to this generalized model and find that in the cosmological context, different choices of Lagrangian density will apparently result in different phases of the Universe. By carefully choosing the variables, we prove that there is an attractor solution to describe the late time accelerating universe when the modified gravity is chosen in a simple power-law form of the curvature scalar. We further examine the temperature evolution based on the thermodynamic understanding of the model. Confronting the model with supernova type Ia data sets, we find that the nonminimally coupled theory of gravity is a viable model to describe the late time Universe acceleration.

2004.08212

Stateful Premise Selection by Recurrent Neural Networks

In this work, we develop a new learning-based method for selecting facts (premises) when proving new goals over large formal libraries. Unlike previous methods that choose sets of facts independently of each other by their rank, the new method uses the notion of \emph{state} that is updated each time a choice of a fact is made. Our stateful architecture is based on recurrent neural networks which have been recently very successful in stateful tasks such as language translation. The new method is combined with data augmentation techniques, evaluated in several ways on a standard large-theory benchmark, and compared to state-of-the-art premise approach based on gradient boosted trees. It is shown to perform significantly better and to solve many new problems.

Our work addresses two important issues with recurrent neural networks: (1) they are over-parameterized, and (2) the recurrence matrix is ill-conditioned. The former increases the sample complexity of learning and the training time. The latter causes the vanishing and exploding gradient problem. We present a flexible recurrent neural network model called Kronecker Recurrent Units (KRU). KRU achieves parameter efficiency in RNNs through a Kronecker factored recurrent matrix. It overcomes the ill-conditioning of the recurrent matrix by enforcing soft unitary constraints on the factors. Thanks to the small dimensionality of the factors, maintaining these constraints is computationally efficient. Our experimental results on seven standard data-sets reveal that KRU can reduce the number of parameters by three orders of magnitude in the recurrent weight matrix compared to the existing recurrent models, without trading the statistical performance. These results in particular show that while there are advantages in having a high dimensional recurrent space, the capacity of the recurrent part of the model can be dramatically reduced.

2004.08214

Accelerating bianchi type dark energy cosmological model with cosmic string in f(T) Gravity

In this paper, we have investigated some features of anisotropic accelerating Bianchi type-I cosmological model in the presence of two non-interacting fluids, i.e. one usual string and other dark energy fluid towards the gravitational field equations for the linear form of $f(T)$ gravity, where \textit{T }be the torsion. To achieve a physically realistic solution of the field equations, we have considered an exact matter dominated volumetric power law expansion. The aspects of derived model are discussed with the help of solution and make the model at late times turn out to be flat Universe. Also, observed that the model has initially non-singular and stable while for whole expansion it is unstable. At an initial expansion $0.01\le t\le 0.05$ when the Universe start to expand to infinite expansion $t>1.73$ (break away from for small interval of cosmic time $0.06\le t\le 1.73$), there is a matter dominated era while for $0.06\le t\le 1.73$ it approaches the quintessence region. Also, the string exists in the early universe which is in agreement with constraints of CMBR data but with the expansion the string phase of the Universe disappears, i.e. we have an anisotropic fluid of particles.

We review on the viability of $F(R)$-gravity. We show that recent cosmic acceleration, radiation/matter-dominated epoch and inflation could be realized in the framework of $F(R)$-gravity in the unified way. For some classes of $F(R)$-gravity, the correction to the Newton law is extremely small and there is no so-called matter instability (the very heavy positive mass for additional scalar degree of freedom is generated). The reconstruction program in modified gravity is also reviewed and it is demonstrated that {\it any} time-evolution of the universe expansion could be realized in $F(R)$-gravity. Special attention is paid to modified gravity which unifies inflation with cosmic acceleration and passes local tests. It turns out that such a theory may describe also dark matter.

2004.08215

Advanced data analysis procedure for hard x-ray resonant magnetic reflectivity discussed for Pt thin film samples of various complexity

X-ray resonant magnetic reflectivity (XRMR) is a powerful method to determine the optical, structural and magnetic depth profiles of a variety of thin films. Here, we investigate samples of different complexity all measured at the Pt L$_3$ absorption edge to determine the optimal procedure for the analysis of the experimental XRMR curves, especially for nontrivial bi- and multilayer samples that include differently bonded Pt from layer to layer. The software tool ReMagX is used to fit these data and model the magnetooptic depth profiles based on a highly adaptable layer stack which is modified to be a more precise and physically consistent representation of the real multilayer system. Various fitting algorithms, iterative optimization approaches and a detailed analysis of the asymmetry ratio features as well as $\chi^2$ (goodness of fit) landscapes are utilized to improve the agreement between measurements and simulations. We present a step-by-step analysis procedure tailored to the Pt thin film systems to take advantage of the excellent magnetic sensitivity and depth resolution of XRMR.

We have examined whether the Co ions crystallographically substitute on the Ti sites in rutile and anatase Ti_{1-x}$Co$_{x}$O$_{2-delta}$ thin films that exhibit room-temperature ferromagnetism. Intensities of the x-ray Bragg reflection from the films were measured around the $K$-absorption-edge of Co. If the Co ions randomly substitute on the Ti sites, the intensity should exhibit an anomaly due to the anomalous dispersion of the atomic scattering factor of Co. However, none of the anatase and rutile samples did exhibit an anomaly, unambiguously showing that the Co ions in Ti$_{1-x}$Co$_{x}$O$_{2-delta}$ are not exactly located at the Ti sites of TiO$_2$. The absence of the anomaly is probably caused by a significant deformation of the local structure around Co due to the oxygen vacancy. We have applied the same method to paramagnetic Zn$_{1-x}$Co$_{x}$O thin films and obtained direct evidence that the Co ions are indeed substituted on the Zn sites.

2004.08217

Asymptotic Analysis of an Ensemble of Randomly Projected Linear Discriminants

Datasets from the fields of bioinformatics, chemometrics, and face recognition are typically characterized by small samples of high-dimensional data. Among the many variants of linear discriminant analysis that have been proposed in order to rectify the issues associated with classification in such a setting, the classifier in [1], composed of an ensemble of randomly projected linear discriminants, seems especially promising; it is computationally efficient and, with the optimal projection dimension parameter setting, is competitive with the state-of-the-art. In this work, we seek to further understand the behavior of this classifier through asymptotic analysis. Under the assumption of a growth regime in which the dataset and projection dimensions grow at constant rates to each other, we use random matrix theory to derive asymptotic misclassification probabilities showing the effect of the ensemble as a regularization of the data sample covariance matrix. The asymptotic errors further help to identify situations in which the ensemble offers a performance advantage. We also develop a consistent estimator of the misclassification probability as an alternative to the computationally-costly cross-validation estimator, which is conventionally used for parameter tuning. Finally, we demonstrate the use of our estimator for tuning the projection dimension on both real and synthetic data.

This article carries out a large dimensional analysis of standard regularized discriminant analysis classifiers designed on the assumption that data arise from a Gaussian mixture model with different means and covariances. The analysis relies on fundamental results from random matrix theory (RMT) when both the number of features and the cardinality of the training data within each class grow large at the same pace. Under mild assumptions, we show that the asymptotic classification error approaches a deterministic quantity that depends only on the means and covariances associated with each class as well as the problem dimensions. Such a result permits a better understanding of the performance of regularized discriminant analsysis, in practical large but finite dimensions, and can be used to determine and pre-estimate the optimal regularization parameter that minimizes the misclassification error probability. Despite being theoretically valid only for Gaussian data, our findings are shown to yield a high accuracy in predicting the performances achieved with real data sets drawn from the popular USPS data base, thereby making an interesting connection between theory and practice.

2004.08218

Inertial modes of a freely rotating ellipsoidal planet and their relation to nutations

We compute the inertial modes of a freely rotating two-layer planetary model with an ellipsoidal inviscid fluid core and a perfectly rigid mantle. We derive analytical formulae for the frequencies of the Free Core Nutation (FCN) and Chandler Wobble (CW) which are valid to all orders in the core and mantle ellipticity and we show how the FCN and CW are the direct generalisation of the purely fluid Spin-Over mode (SO) and of the Eulerian Wobble (EW) to the case where the mantle can oscillate freely around a state of steady rotation. Through a numerical computation for an axisymmetric (oblate spheroidal) planet, we demonstrate that all other inertial modes of the steadily rotating fluid core are also free modes of the freely rotating two-layer planet.

Context. Ceres is the most massive body of the asteroid belt and contains about 25 wt.% (weight percent) of water. Understanding its thermal evolution and assessing its current state are major goals of the Dawn Mission. Constraints on internal structure can be inferred from various observations. Especially, detailed knowledge of the rotational motion can help constrain the mass distribution inside the body, which in turn can lead to information on its geophysical history. Aims. We investigate the signature of the interior on the rotational motion of Ceres and discuss possible future measurements performed by the spacecraft Dawn that will help to constrain Ceres' internal structure. Methods. We compute the polar motion, precession-nutation, and length-of-day variations. We estimate the amplitudes of the rigid and non-rigid response for these various motions for models of Ceres interior constrained by recent shape data and surface properties. Results. As a general result, the amplitudes of oscillations in the rotation appear to be small, and their determination from spaceborne techniques will be challenging. For example, the amplitudes of the semi-annual and annual nutations are around ~364 and ~140 milli-arcseconds, and they show little variation within the parametric space of interior models envisioned for Ceres. This, combined with the very long-period of the precession motion, requires very precise measurements. We also estimate the timescale for Ceres' orientation to relax to a generalized Cassini State, and we find that the tidal dissipation within that object was probably too small to drive any significant damping of its obliquity since formation. However, combining the shape and gravity observations by Dawn offers the prospect to identify departures of non-hydrostaticity at the global and regional scale, which will be instrumental in constraining Ceres' past and current thermal state. We also discuss the existence of a possible Chandler mode in the rotational motion of Ceres, whose potential excitation by endogenic and/or exogenic processes may help detect the presence of liquid reservoirs within the asteroid.

2004.08219

Exploratory Test Agents for Stateful Software Systems

The adequate testing of stateful software systems is a hard and costly activity. Failures that result from complex stateful interactions can be of high impact, and it can be hard to replicate failures resulting from erroneous stateful interactions. Addressing this problem in an automatic way would save cost and time and increase the quality of software systems in the industry. In this paper, we propose an approach that uses agents to explore software systems with the intention to find faults and gain knowledge.

Component-based software development has posed a serious challenge to system verification since externally-obtained components could be a new source of system failures. This issue can not be completely solved by either model-checking or traditional software testing techniques alone due to several reasons: 1) externally obtained components are usually unspecified/partially specified; 2)it is generally difficult to establish an adequacy criteria for testing a component; 3)components may be used to dynamically upgrade a system. This paper introduces a new approach (called {\em model-checking driven black-box testing}) that combines model-checking with traditional black-box software testing to tackle the problem in a complete, sound, and automatic way. The idea is to, with respect to some requirement (expressed in CTL or LTL) about the system, use model-checking techniques to derive a condition (expressed in communication graphs) for an unspecified component such that the system satisfies the requirement iff the condition is satisfied by the component, and which can be established by testing the component with test cases generated from the condition on-the-fly. In this paper, we present model-checking driven black-box testing algorithms to handle both CTL and LTL requirements. We also illustrate the idea through some examples.

2004.08220

Reconciling emergences: An information-theoretic approach to identify causal emergence in multivariate data

The broad concept of emergence is instrumental in various of the most challenging open scientific questions -- yet, few quantitative theories of what constitutes emergent phenomena have been proposed. This article introduces a formal theory of causal emergence in multivariate systems, which studies the relationship between the dynamics of parts of a system and macroscopic features of interest. Our theory provides a quantitative definition of downward causation, and introduces a complementary modality of emergent behaviour -- which we refer to as causal decoupling. Moreover, the theory allows practical criteria that can be efficiently calculated in large systems, making our framework applicable in a range of scenarios of practical interest. We illustrate our findings in a number of case studies, including Conway's Game of Life, Reynolds' flocking model, and neural activity as measured by electrocorticography.

Background: Identification of causal SNPs in most genome wide association studies relies on approaches that consider each SNP individually. However, there is a strong correlation structure among SNPs that need to be taken into account. Hence, increasingly modern computationally expensive regression methods are employed for SNP selection that consider all markers simultaneously and thus incorporate dependencies among SNPs. Results: We develop a novel multivariate algorithm for large scale SNP selection using CAR score regression, a promising new approach for prioritizing biomarkers. Specifically, we propose a computationally efficient procedure for shrinkage estimation of CAR scores from high-dimensional data. Subsequently, we conduct a comprehensive comparison study including five advanced regression approaches (boosting, lasso, NEG, MCP, and CAR score) and a univariate approach (marginal correlation) to determine the effectiveness in finding true causal SNPs. Conclusions: Simultaneous SNP selection is a challenging task. We demonstrate that our CAR score-based algorithm consistently outperforms all competing approaches, both uni- and multivariate, in terms of correctly recovered causal SNPs and SNP ranking. An R package implementing the approach as well as R code to reproduce the complete study presented here is available from http://strimmerlab.org/software/care/ .

2004.08221

Contact rate epidemic control of COVID-19: an equilibrium view

We consider the control of the COVID-19 pandemic, modeled by a standard SIR compartmental model. The control of the epidemic is induced by the aggregation of individuals' decisions to limit their social interactions: on one side, when the epidemic is ongoing, an individual is encouraged to diminish his/her contact rate in order to avoid getting infected, but, on the other side, this effort comes at a social cost. If each individual lowers his/her contact rate, the epidemic vanishes faster but the effort cost may be high. A Mean Field Nash equilibrium at the population level is formed, resulting in a lower effective transmission rate of the virus. However, it is not clear that the individual's interest aligns with that of the society. We prove that the equilibrium exists and compute it numerically. The equilibrium selects a sub-optimal solution in comparison to the societal optimum (a centralized decision respected fully by all individuals), meaning that the cost of anarchy is strictly positive. We provide numerical examples and a sensitivity analysis. We show that the divergence between the individual and societal strategies happens after the epidemic peak but while significant propagation is still underway.

The outbreak of the novel coronavirus, officially declared a global pandemic, has a severe impact on our daily lives. As of this writing there are approximately 197,188 confirmed cases of which 80,881 are in "Mainland China" with 7,949 deaths, a mortality rate of 3.4%. In order to support radiologists in this overwhelming challenge, we develop a deep learning based algorithm that can detect, localize and quantify severity of COVID-19 manifestation from chest CT scans. The algorithm is comprised of a pipeline of image processing algorithms which includes lung segmentation, 2D slice classification and fine grain localization. In order to further understand the manifestations of the disease, we perform unsupervised clustering of abnormal slices. We present our results on a dataset comprised of 110 confirmed COVID-19 patients from Zhejiang province, China.

2004.08222

Learning to Predict Context-adaptive Convolution for Semantic Segmentation

Long-range contextual information is essential for achieving high-performance semantic segmentation. Previous feature re-weighting methods demonstrate that using global context for re-weighting feature channels can effectively improve the accuracy of semantic segmentation. However, the globally-sharing feature re-weighting vector might not be optimal for regions of different classes in the input image. In this paper, we propose a Context-adaptive Convolution Network (CaC-Net) to predict a spatially-varying feature weighting vector for each spatial location of the semantic feature maps. In CaC-Net, a set of context-adaptive convolution kernels are predicted from the global contextual information in a parameter-efficient manner. When used for convolution with the semantic feature maps, the predicted convolutional kernels can generate the spatially-varying feature weighting factors capturing both global and local contextual information. Comprehensive experimental results show that our CaC-Net achieves superior segmentation performance on three public datasets, PASCAL Context, PASCAL VOC 2012 and ADE20K.

Computer-aided diagnosis system for diffuse lung diseases (DLDs) is necessary for the objective assessment of the lung diseases. In this paper, we develop semantic segmentation model for 5 kinds of DLDs. DLDs considered in this work are consolidation, ground glass opacity, honeycombing, emphysema, and normal. Convolutional neural network (CNN) is one of the most promising technique for semantic segmentation among machine learning algorithms. While creating annotated dataset for semantic segmentation is laborious and time consuming, creating partially annotated dataset, in which only one chosen class is annotated for each image, is easier since annotators only need to focus on one class at a time during the annotation task. In this paper, we propose a new weak supervision technique that effectively utilizes partially annotated dataset. The experiments using partially annotated dataset composed 372 CT images demonstrated that our proposed technique significantly improved segmentation accuracy.

2004.08224

On the nonexistence of harmonic and bi-harmonic maps

In this paper, we study the existence of harmonic and bi-harmonic maps into Riemannian manifolds admitting a conformal vector field, or a nontrivial Ricci solitons.

The tension equation for a mapping $f:{\mathbb C}\to {\mathbb C}$ is the nonlinear second order equation \[ \Delta f +\varphi(f) f_z f_{\bar z} = 0\] Solutions are "harmonic" mappings. Here we give a complete description of the solution space of mappings of degree 1 to this equation when $\varphi$ is entire. Each solution is a quasiconformal surjection and when the set of normalised solutions is endowed with the Teichm\"uller metric, the solution space is isometric to the hyperbolic plane. More generally, for harmonic mappings $f:\Omega \to (\tilde{\Omega},\rho)$ between domains in ${\mathbb C}$, with $\rho(w)|dw|$ defining a flat metric we stablish a very strong maximum principle for the distortion - up to multiplicative factor $e^{iv}$, $v$ real and harmonic, the Beltrami coefficient of $f^{-1}$ is quasiregular - and thus open and discrete when nonconstant. This follows from the remarkable fact that the Beltrami coefficient of the inverse of a harmonic mapping itself satisfies a nonlinear homogeneous Beltrami equation.

2004.08225

Unruh-like effects: Effective temperatures along stationary worldlines

We study the temperatures recorded along all classes of stationary, uniformly accelerated worldlines in Minkowski spacetime, namely along (i) linear uniform acceleration, (ii) cusped, (iii) circular, (iv) catenary, and (v) helix worldlines, of which the Unruh temperature is the particular case for linear uniform acceleration. As a measuring device, we employ an Unruh-DeWitt detector, modeled as a qubit that interacts for a long time with a massless Klein-Gordon field in the vacuum state. The temperatures in each case (i) - (v) are functions of up to three invariant quantities: curvature or proper acceleration, $\kappa$, torsion, $b$, and hypertorsion, $\nu$, and except for the case (i), they depend on the transition frequency difference of the detector. We investigate numerically the behavior of the frequency-dependent temperatures for different values of $\kappa$, $b$, and $\nu$ along the stationary worldlines (ii) - (v) and evaluate analytically the regimes where the temperatures recorded along the different worldlines coincide with each other in terms of relevant asymptotic limits for $\kappa$, $b$, or $\nu$, and discuss their physical meaning. We demonstrate that the temperatures in cases (ii) - (v) dip under the Unruh temperature at low frequencies and go above the Unruh temperature for large $|\omega|$. It is our hope that this study will be relevant to the design of experiments seeking to verify the Unruh effect or generalizations thereof.

Surface waves on a stationary flow of water are considered, in a linear model that includes the surface tension of the fluid. The resulting gravity-capillary waves experience a rich array of horizon effects when propagating against the flow. In some cases three horizons (points where the group velocity of the wave reverses) exist for waves with a single laboratory frequency. Some of these effects are familiar in fluid mechanics under the name of wave blocking, but other aspects, in particular waves with negative co-moving frequency and the Hawking effect, were overlooked until surface waves were investigated as examples of analogue gravity [Sch\"utzhold R and Unruh W G 2002 Phys. Rev. D 66 044019]. A comprehensive presentation of the various horizon effects for gravity-capillary waves is given, with emphasis on the deep water/short wavelength case kh>>1 where many analytical results can be derived. A similarity of the state space of the waves to that of a thermodynamic system is pointed out.

2004.08226

Core polarizability of rubidium using spectroscopy of the ng to nh, ni Rydberg transitions

We present a precise measurement of the rubidium ionic core polarizability. The results can be useful for interpreting experiments such as parity violation or black-body radiation shifts in atomic clocks since the ionic core electrons contribute significantly to the total electrical polarizability of rubidium. We report a dipole polarizability $\alpha_d$ = $9.063 \pm 0.007$ $a_0^3$ and quadrupole polarizability $\alpha_q$ = $43 \pm 2$ $a_{0}^{5}$ derived from microwave and radio-frequency spectroscopy measurements of Rydberg states with large angular momentum. By using a relatively low principal quantum number ($17 \leq n \leq 19$) and high angular momentum ($4 \leq \ell \leq 6$), systematic effects are reduced compared to previous experiments. We account for non-adiabatic corrections to the polarizability model, which resolves a previous discrepancy with theory but limits the accuracy of our results.

The electric fields near the heterogeneous metal/dielectric surface of an atom chip were measured using cold atoms. The atomic sensitivity to electric fields was enhanced by exciting the atoms to Rydberg states that are 10^8 times more polarizable than the ground state. We attribute the measured fields to charging of the insulators between the atom chip wires. Surprisingly, it is observed that these fields may be dramatically lowered with appropriate voltage biasing, suggesting configurations for the future development of hybrid quantum systems.

2004.08227

MPLP++: Fast, Parallel Dual Block-Coordinate Ascent for Dense Graphical Models

Dense, discrete Graphical Models with pairwise potentials are a powerful class of models which are employed in state-of-the-art computer vision and bio-imaging applications. This work introduces a new MAP-solver, based on the popular Dual Block-Coordinate Ascent principle. Surprisingly, by making a small change to the low-performing solver, the Max Product Linear Programming (MPLP) algorithm, we derive the new solver MPLP++ that significantly outperforms all existing solvers by a large margin, including the state-of-the-art solver Tree-Reweighted Sequential (TRWS) message-passing algorithm. Additionally, our solver is highly parallel, in contrast to TRWS, which gives a further boost in performance with the proposed GPU and multi-thread CPU implementations. We verify the superiority of our algorithm on dense problems from publicly available benchmarks, as well, as a new benchmark for 6D Object Pose estimation. We also provide an ablation study with respect to graph density.

Massive data sets have radically changed our understanding of how to design efficient algorithms; the streaming paradigm, whether it in terms of number of passes of an external memory algorithm, or the single pass and limited memory of a stream algorithm, appears to be the dominant method for coping with large data. A very different kind of massive computation has had the same effect at the level of the CPU. The most prominent example is that of the computations performed by a graphics card. The operations themselves are very simple, and require very little memory, but require the ability to perform many computations extremely fast and in parallel to whatever degree possible. What has resulted is a stream processor that is highly optimized for stream computations. An intriguing side effect of this is the growing use of a graphics card as a general purpose stream processing engine. In an ever-increasing array of applications, researchers are discovering that performing a computation on a graphics card is far faster than performing it on a CPU, and so are using a GPU as a stream co-processor.

2004.08229

Comparison of pulsar parameters for some anomalous types

The comparative analysis of parameters is carried out for two samples of radio pulsars. Objects of the first sample have periods P > 2 sec, the second is characterized by magnetic fields at the neutron star surface $Bs > 4.4\times10^{13}$ G. The main goal of this analysis is to find some differences between normal pulsars and anomalous X-ray pulsars (AXP) and soft gamma-ray repeaters (SGR) with the similar values of periods and magnetic fields. It is shown that the dependence of period derivatives on periods for pulsars with long periods is $dP/dt \propto P^{1.67}$. This means that we must search for another braking mechanism differing from the magnetic dipole one which gives $dP/dt \propto P^{-1}$ or use some specific distributions of parameters. On the other hand, radio pulsars with strong magnetic fields and AXP/SGR can be described by the magnetic dipole model. Radio pulsars with P > 2 sec have ages on average two orders of magnitude higher than objects with $Bs > 4.4 \times 10^{13}$ G and are located much higher above the Galactic plane. The obtained results show that the main feature of AXP/SGR is the strong magnetic field and not the large period.

Luminosity is an intrinsic property of radio pulsars related to the properties of the magnetospheric plasma and the beam geometry, and inversely proportional to the observing frequency. In traditional models, luminosity has been considered as a function of the spin parameters of pulsars. On the other hand, parameter independent models like power law and lognormal have been also used to fit the observed luminosities. Some of the older studies on pulsar luminosities neglected observational biases, but all of the recent studies tried to model observational effects as accurately as possible. Luminosities of pulsars in globular clusters and in the Galactic disk have been studied separately. Older studies concluded that these two categories of pulsars have different luminosity distributions, but the most recent study concluded that those are the same. This article reviews all significant works on pulsar luminosities and discusses open questions.

2004.08231

Characterizing Quantum Correlations In Fixed Input $n$-Local Network Scenario

Contrary to Bell scenario, quantum nonlocality can be exploited even when all the parties do not have freedom to select inputs randomly. Such manifestation of nonlocality is possible in networks involving independent sources. One can utilize such a feature of quantum networks for purpose of entanglement detection of bipartite quantum states. In this context, we characterize correlations simulated in networks involving finite number of sources generating quantum states when some parties perform fixed measurement. Beyond bipartite entanglement, we enquire the same for networks involving sources now generating pure tripartite quantum states. Interestingly, here also randomness in input selection is not necessary for every party to generate nonlocal correlation.

Non-recurrent congestion is a major problem in traffic networks that causes unexpected delays during travels. In such a scenario, it is preferable to use adaptive paths or policies where next link decisions on reaching junctions are continuously adapted based on the information gained with time. In this paper, we study a traffic assignment problem in stochastic time-dependent networks. The problem is modeled as a fixed-point problem and existence of the equilibrium solution is discussed. We iteratively solve the problem using the method of successive averages (MSA). A novel network loading model inspired from Link transmission model is developed that accepts policies as inputs for solving the problem. This network loading model is different from the existing network loading models that take predefined paths for input flows. We demonstrate through numerical tests that solving traffic assignment problem with the proposed loading modeling scheme is more efficient as compared to solving the problem using path-based network loading models.

2004.08232

Representation of two by two matrices over perfect fields by a diagonal quadratic form

Let F be a perfect field. Then the diagonal quadratic form $a_iX_i^2$ over $F$ is universal over $M_2(F)$ if and only if atleast two of the $a_i$ are non-zero.

We introduce the notion of the full quiver of a representation of an algebra, which is a cover of the (classical) quiver, but which captures properties of the representation itself. Gluing of vertices and of arrows enables one to study subtle combinatorial aspects of algebras which are lost in the classical quiver. Full quivers of representations apply especially well to \Zcd\ algebras, which have properties very like those of finite dimensional algebras over fields. By choosing the representation appropriately, one can restrict the gluing to two main types: {\it Frobenius} (along the diagonal) and, more generally {\it proportional} Frobenius gluing (above the diagonal), and our main result is that any representable algebra has a faithful representation described completely by such a full quiver. Further reductions are considered, which bear on the polynomial identities.

2004.08233

Doubly-Charged Bileptons at the LHC

We discuss a minimal and simple extension of the standard electroweak model and discuss its uniqueness properties. Its most readily testable prediction is the existence of $|Q|= |L|= 2$ gauge bosons, bileptons, which decay into like-sign lepton pairs. We discuss how a discovery could be made at the LHC.

In this piece of work, the nonstandard Higgs boson masses have been constrained by restrictions arising from Higgs diphoton decay width, stability, perturbativity, unitarity and electroweak T-parameter in the framework of the Higgs triplet model. As a matter of fact the loop induced Higgs decays will now obtain additional contributions from singly and doubly charged Higgs bosons. In simulations of pair production of $H^{++}$ it is assumed that the production channel $q\overline{q}\rightarrow \gamma^{*},Z^{*}\rightarrow H^{++}H^{--}$ is the only mechanism. Whereas there may be other channels of production of doubly charged Higgs bosons as for instance from the decay of singly charged Higgs bosons. The effect of this additional sources of production of the contributing non-standard charged particles to the loop induced Higgs decays has been studied in this paper.

2004.08235

Multipolar, Polarization Shaped High Harmonic Generation by Intense Vector Beams

High harmonic generation (HHG) is a manifestation of the strongly nonlinear response of matter to intense laser fields and has, as the basis for coherent XUV sources a variety of applications. Recently, HHG from atoms in a phase and polarization structured laser was demonstrated and interpreted based on the transverse electric field component of the driving pulse. Here we point out that as dictated by Maxwell equations, such fields have a longitudinal component which in general has a fundamental influence on the charge dynamics. For instance, its interplay with the transversal field component enables endowing the emitted radiation locally with circular polarization and a defined polarity. It is shown that the time-dependent Stokes parameters defining the polarization state of HHG can be tuned by varying the waist of the driving field which in turn, changes the ratio between the longitudinal and transverse electric-field components of the driving laser. In addition, employing a multipole expansion of the produced harmonics exposes the specific multipolar character and the relation to the spatial structure of the driving field polarization states. The scheme proposed here allows a full polarization control of the emitted harmonics by only one driving laser. A tighter focusing of the driving pulse renders possible the emission of harmonics with both even and odd spatial symmetry. The underlying mechanism is due to the fundamental interplay between the transverse and longitudinal components of the laser's electromagnetic vector potential. The ratio between those components is controllable by just focusing the laser spot, pointing to an accessible tool for polarization and polarity control of the high harmonics.

Nonlocal nonlinear Schroedinger-type equation is derived as a model to describe paraxial light propagation in nonlinear media with different `degrees' of nonlocality. High frequency limit of this equation is studied under specific assumptions of Cole-Cole dispersion law and a slow dependence along propagating direction. Phase equations are integrable and they correspond to dispersionless limit of Veselov-Novikov hierarchy. Analysis of compatibility among intensity law (dependence of intensity on the refractive index) and high frequency limit of Poynting vector conservation law reveals the existence of singular wavefronts. It is shown that beams features depend critically on the orientation properties of quasiconformal mappings of the plane. Another class of wavefronts, whatever is intensity law, is provided by harmonic minimal surfaces. Illustrative example is given by helicoid surface. Compatibility with first and third degree nonlocal perturbations and explicit solutions are also discussed.

2004.08236

Chirality Through Classical Physics

Chirality, or handedness, is a topic that is common in biology and chemistry, yet is rarely discussed in physics courses. We provide a way of introducing the topic in classical physics, and demonstrate the merits of its inclusion - such as a simple way to visually introduce the concept of symmetries in physical law - along with giving some simple proofs using only basic matrix operations, thereby avoiding the full formalism of the three-dimensional point group.

We calculate the helicity and chirality effects experienced by a spin-1/2 particle subjected to classical electromagnetic and gravitational fields. The helicity evolution is then determined in the non-relativistic, relativistic, and ultra-relativistic regimes. We find that inertia-gravitation can distinguish between helicity and chirality. Helicity is not conserved, in general, even when the particles are massless. In this case, however, the inertial fields can hardly be applied to the fermions.

2004.08237

CAggNet: Crossing Aggregation Network for Medical Image Segmentation

In this paper, we present Crossing Aggregation Network (CAggNet), a novel densely connected semantic segmentation method for medical image analysis. The crossing aggregation network absorbs the idea of deep layer aggregation and makes significant innovations in layer connection and semantic information fusion. In this architecture, the traditional skip-connection structure of general U-Net is replaced by aggregations of multilevel down-sampling and up-sampling layers. This enables the network to fuse information interactively flows at different levels of layers in semantic segmentation. It also introduces weighted aggregation module to aggregate multi-scale output information. We have evaluated and compared our CAggNet with several advanced U-Net based methods in two public medical image datasets, including the 2018 Data Science Bowl nuclei detection dataset and the 2015 MICCAI gland segmentation competition dataset. Experimental results indicate that CAggNet improves medical object recognition and achieves a more accurate and efficient segmentation compared to existing improved U-Net and UNet++ structure.

Inspired by classic generative adversarial networks (GAN), we propose a novel end-to-end adversarial neural network, called SegAN, for the task of medical image segmentation. Since image segmentation requires dense, pixel-level labeling, the single scalar real/fake output of a classic GAN's discriminator may be ineffective in producing stable and sufficient gradient feedback to the networks. Instead, we use a fully convolutional neural network as the segmentor to generate segmentation label maps, and propose a novel adversarial critic network with a multi-scale $L_1$ loss function to force the critic and segmentor to learn both global and local features that capture long- and short-range spatial relationships between pixels. In our SegAN framework, the segmentor and critic networks are trained in an alternating fashion in a min-max game: The critic takes as input a pair of images, (original_image $*$ predicted_label_map, original_image $*$ ground_truth_label_map), and then is trained by maximizing a multi-scale loss function; The segmentor is trained with only gradients passed along by the critic, with the aim to minimize the multi-scale loss function. We show that such a SegAN framework is more effective and stable for the segmentation task, and it leads to better performance than the state-of-the-art U-net segmentation method. We tested our SegAN method using datasets from the MICCAI BRATS brain tumor segmentation challenge. Extensive experimental results demonstrate the effectiveness of the proposed SegAN with multi-scale loss: on BRATS 2013 SegAN gives performance comparable to the state-of-the-art for whole tumor and tumor core segmentation while achieves better precision and sensitivity for Gd-enhance tumor core segmentation; on BRATS 2015 SegAN achieves better performance than the state-of-the-art in both dice score and precision.

2004.08238

Novel Binary-Addition Tree Algorithm (BAT) for Binary-State Network Reliability Problem

Network structures and models have been widely adopted, e.g., for Internet of Things, wireless sensor networks, smart grids, transportation networks, communication networks, social networks, and computer grid systems. Network reliability is an effective and popular technique to estimate the probability that the network is still functioning. Networks composed of binary-state (e.g., working or failed) components (arcs and/or nodes) are called binary-state networks. The binary-state network is the fundamental type of network; thus, there is always a need for a more efficient algorithm to calculate the network reliability. Thus, a novel binary-addition tree (BAT) algorithm that employs binary addition for finding all the possible state vectors and the path-based layered-search algorithm for filtering out all the connected vectors is proposed for calculating the binary-state network reliability. According to the time complexity and numerical examples, the efficiency of the proposed BAT is higher than those of traditional algorithms for solving the binary-state network reliability problem.

Deep learning has made significant improvements at many image processing tasks in recent years, such as image classification, object recognition and object detection. Convolutional neural networks (CNN), which is a popular deep learning architecture designed to process data in multiple array form, show great success to almost all detection \& recognition problems and computer vision tasks. However, the number of parameters in a CNN is too high such that the computers require more energy and larger memory size. In order to solve this problem, we propose a novel energy efficient model Binary Weight and Hadamard-transformed Image Network (BWHIN), which is a combination of Binary Weight Network (BWN) and Hadamard-transformed Image Network (HIN). It is observed that energy efficiency is achieved with a slight sacrifice at classification accuracy. Among all energy efficient networks, our novel ensemble model outperforms other energy efficient models.

2004.08239

Three-dimensional Navier-Stokes Equations. Existence and smoothness of solutions on maximal time intervals

The existence and smoothness of solution to the three-dimensional Navier-Stokes equations on maximal time intervals is proved in this paper. This is a part of the Clay Millennium Prize Problem. The nonexistence of a blow-up solution for small data is proved in this paper.

The aim of this paper is to prove the strong convergence of the solutions to a vector-BGK model under the diffusive scaling to the incompressible Navier-Stokes equations on the two-dimensional torus. This result holds in any interval of time $[0, T]$, with $T>0$. We also provide the global in time uniform boundedness of the solutions to the approximating system. Our argument is based on the use of local in time $H^s$-estimates for the model, established in a previous work, combined with the $L^2$-relative entropy estimate and the interpolation properties of the Sobolev spaces.

2004.08240

Designing a NISQ reservoir with maximal memory capacity for volatility forecasting

Quantitative risk management, particularly volatility forecasting, is critically important to traders, portfolio managers as well as policy makers. In this paper, we applied quantum reservoir computing for forecasting VIX (the CBOE volatility index), a highly non-linear and memory intensive `real-life' signal that is driven by market dynamics and trader psychology and cannot be expressed by a deterministic equation. As a first step, we lay out the systematic design considerations for using a NISQ reservoir as a computing engine (which should be useful for practitioners). We then show how to experimentally evaluate the memory capacity of various reservoir topologies (using IBM-Q's Rochester device) to identify the configuration with maximum memory capacity. Once the optimal design is selected, the forecast is produced by a linear combination of the average spin of a 6-qubit quantum register trained using VIX and SPX data from year 1990 onwards. We test the forecast performance over the sub-prime mortgage crisis period (Dec 2007 - Jun 2009). Our results show a remarkable ability to predict the volatility during the Great Recession using today's NISQs.

A new class of non-homogeneous state-affine systems is introduced for use in reservoir computing. Sufficient conditions are identified that guarantee first, that the associated reservoir computers with linear readouts are causal, time-invariant, and satisfy the fading memory property and second, that a subset of this class is universal in the category of fading memory filters with stochastic almost surely uniformly bounded inputs. This means that any discrete-time filter that satisfies the fading memory property with random inputs of that type can be uniformly approximated by elements in the non-homogeneous state-affine family.

2004.08241

SICs: Some explanations

The problem of constructing maximal equiangular tight frames or SICs was raised by Zauner in 1998. Four years ago it was realized that the problem is closely connected to a major open problem in number theory. We discuss why such a connection was perhaps to be expected, and give a simplified sketch of some developments that have taken place in the past four years. The aim, so far unfulfilled, is to prove existence of SICs in an infinite sequence of dimensions.

The reliability of power semiconductor switches is important when considering their vital role in power electronic converters for downhole subsea applications. Respect to technology advancements in material sciences, power MOSFETs with wide band gap materials have been proposed such as silicon carbide (SiC) and gallium nitride (GaN) as an alternative to existing silicon (Si) based MOSFETs and IGBTs. However, reliability analysis should be performed before substituting SiC-MOSFETs in the place of existing Si-MOSFETs and IGBTs. Due to costly equipment of experimental test setup for accelerated life test, a good reliable and precise simulation-based test bench should be used to test the life test procedure before implementing actual hardware. Therefore, this paper introduces a power cycle (PC) test bench for accelerated life testing for reliability assessment of SiC-MOSFET in harsh offshore environment. The introduced test bench is a simulation-based of power switch in SimScape and LTspice and has been validated with datasheet of 1.2 kV SiC-MOSFET, CAS300M12BM2 by CREE. Preliminary hardware circuits are also shown for further experimental tests. The captured data from the Device-Under-Test (DUT) in different ambient temperatures are envisioned and provide critical information about the failure mechanisms and lifetime characteristics of power devices. The provided lifetime characteristics data of SiC-MOSFET can be used to statistically estimate the Remaining-Useful-Lifetime (RUL) of component in a real application such as downhole motor drives.

2004.08242

Spectral Modeling of Charge Exchange in the Central Region of M51

Charge exchange (CX) emission reveals the significant interaction between neutral and ionized interstellar medium (ISM) components of the dense, multiphase, circumnuclear region of a galaxy. We use a model including a thermal and a CX components to describe the high-resolution XMM-Newton/RGS spectrum of the diffuse emission in the central region of M51. Representative signatures of CX emission -- especially the prominent OVII forbidden line and the excess emission in the OVIII Ly$\gamma$ lines -- can be well explained by the model. Combined with the Chandra images in the OVIII and the OVII bands, we find the soft X-ray emission is dominated by the jet-driven outflow and its interaction with the ambient neutral material. The jet-driven outflow itself is likely a thermal plasma of $\sim 0.59$ keV, with mostly sub-solar abundances. It runs into the ambient neutral gas, and produces significant CX emission that accounts for one-fifth of the diffuse X-ray emission in the 7--28 {\AA} band. The effective interface area in the CX process is one order of magnitude greater than the geometrical surface area of the jet-driven outflow. The tenuous outflow driven by the nuclear star formation may also contribute a small portion to both the diffuse thermal and CX emission. The photoionization by the active galactic nuclei (AGNs) and the resonance scattering by the hot gas itself are disfavored, though the effects from past AGN events may not be ruled out.

Strong shocks propagating into a partially ionized medium are often associated with optical Balmer lines. This emission is due to impact excitation of neutral hydrogen by hot protons and electrons in the shocked gas. The structure of such Balmer-dominated shocks has been computed in a previous paper (Blasi et al. 2012), where the distribution function of neutral particles was derived from the appropriate Boltzmann equation including coupling with ions and electrons through charge exchange and ionization. This calculation showed how the presence of neutrals can significantly modify the shock structure through the formation of a "neutral-induced" precursor ahead of the shock. Here we follow up on our previous work and investigate the properties of the resulting Balmer emission, with the aim of using the observed radiation as a diagnostic tool for shock parameters. Our main focus is on Supernova Remnant shocks, and we find that, for typical parameters, the H{\alpha} emission typically has a three-component spectral profile, where: 1) a narrow component originates from upstream cold hydrogen atoms, 2) a broad component comes from hydrogen atoms that have undergone charge exchange with shocked protons downstream of the shock, and 3) an intermediate component is due to hydrogen atoms that have undergone charge exchange with warm protons in the neutral-induced precursor. The width of the intermediate line reflects the temperature in the precursor, while the width of the narrow one is left unaltered by the precursor. In addition, we show that the profiles of both the intermediate and broad components generally depart from a thermal distribution. Finally, we show that a significant amount of Balmer emission can be produced in the precursor region if efficient electron heating takes place.

2004.08243

Geometry-aware Domain Adaptation for Unsupervised Alignment of Word Embeddings

We propose a novel manifold based geometric approach for learning unsupervised alignment of word embeddings between the source and the target languages. Our approach formulates the alignment learning problem as a domain adaptation problem over the manifold of doubly stochastic matrices. This viewpoint arises from the aim to align the second order information of the two language spaces. The rich geometry of the doubly stochastic manifold allows to employ efficient Riemannian conjugate gradient algorithm for the proposed formulation. Empirically, the proposed approach outperforms state-of-the-art optimal transport based approach on the bilingual lexicon induction task across several language pairs. The performance improvement is more significant for distant language pairs.

Can neural networks learn to compare graphs without feature engineering? In this paper, we show that it is possible to learn representations for graph similarity with neither domain knowledge nor supervision (i.e.\ feature engineering or labeled graphs). We propose Deep Divergence Graph Kernels, an unsupervised method for learning representations over graphs that encodes a relaxed notion of graph isomorphism. Our method consists of three parts. First, we learn an encoder for each anchor graph to capture its structure. Second, for each pair of graphs, we train a cross-graph attention network which uses the node representations of an anchor graph to reconstruct another graph. This approach, which we call isomorphism attention, captures how well the representations of one graph can encode another. We use the attention-augmented encoder's predictions to define a divergence score for each pair of graphs. Finally, we construct an embedding space for all graphs using these pair-wise divergence scores. Unlike previous work, much of which relies on 1) supervision, 2) domain specific knowledge (e.g. a reliance on Weisfeiler-Lehman kernels), and 3) known node alignment, our unsupervised method jointly learns node representations, graph representations, and an attention-based alignment between graphs. Our experimental results show that Deep Divergence Graph Kernels can learn an unsupervised alignment between graphs, and that the learned representations achieve competitive results when used as features on a number of challenging graph classification tasks. Furthermore, we illustrate how the learned attention allows insight into the the alignment of sub-structures across graphs.

2004.08244

The 2-rank of the class group of some real cyclic quartic number fields

In this paper, we investigate the 2-rank of the class group of some real cyclic quartic number fields. Precisely, we consider the case where the quadratic subfield is Q(\sqrt{l}) with l congruent to 5 modulo 8 is a prime.

We give the class of finite groups which arise as the permutation groups of cyclic codes over finite fields. Furthermore, we extend the results of Brand and Huffman et al. and we find the properties of the set of permutations by which two cyclic codes of length p^r can be equivalent. We also find the set of permutations by which two quasi-cyclic codes can be equivalent.

2004.08245

Quality Prediction on Deep Generative Images

In recent years, deep neural networks have been utilized in a wide variety of applications including image generation. In particular, generative adversarial networks (GANs) are able to produce highly realistic pictures as part of tasks such as image compression. As with standard compression, it is desirable to be able to automatically assess the perceptual quality of generative images to monitor and control the encode process. However, existing image quality algorithms are ineffective on GAN generated content, especially on textured regions and at high compressions. Here we propose a new naturalness-based image quality predictor for generative images. Our new GAN picture quality predictor is built using a multi-stage parallel boosting system based on structural similarity features and measurements of statistical similarity. To enable model development and testing, we also constructed a subjective GAN image quality database containing (distorted) GAN images and collected human opinions of them. Our experimental results indicate that our proposed GAN IQA model delivers superior quality predictions on the generative image datasets, as well as on traditional image quality datasets.

The field of deep generative modeling is dominated by generative adversarial networks (GANs). However, the training of GANs often lacks stability, fails to converge, and suffers from model collapse. It takes an assortment of tricks to solve these problems, which may be difficult to understand for those seeking to apply generative modeling. Instead, we propose two novel generative autoencoders, AE-OTtrans and AE-OTgen, which rely on optimal transport instead of adversarial training. AE-OTtrans and AEOTgen, unlike VAE and WAE, preserve the manifold of the data; they do not force the latent distribution to match a normal distribution, resulting in greater quality images. AEOTtrans and AE-OTgen also produce images of higher diversity compared to their predecessor, AE-OT. We show that AE-OTtrans and AE-OTgen surpass GANs in the MNIST and FashionMNIST datasets. Furthermore, We show that AE-OTtrans and AE-OTgen do state of the art on the MNIST, FashionMNIST, and CelebA image sets comapred to other non-adversarial generative models.

2004.08246

Res-CR-Net, a residual network with a novel architecture optimized for the semantic segmentation of microscopy images

Deep Neural Networks (DNN) have been widely used to carry out segmentation tasks in both electron and light microscopy. Most DNNs developed for this purpose are based on some variation of the encoder-decoder type U-Net architecture, in combination with residual blocks to increase ease of training and resilience to gradient degradation. Here we introduce Res-CR-Net, a type of DNN that features residual blocks with either a bundle of separable atrous convolutions with different dilation rates or a convolutional LSTM. The number of filters used in each residual block and the number of blocks are the only hyperparameters that need to be modified in order to optimize the network training for a variety of different microscopy images.

We propose semantic region-adaptive normalization (SEAN), a simple but effective building block for Generative Adversarial Networks conditioned on segmentation masks that describe the semantic regions in the desired output image. Using SEAN normalization, we can build a network architecture that can control the style of each semantic region individually, e.g., we can specify one style reference image per region. SEAN is better suited to encode, transfer, and synthesize style than the best previous method in terms of reconstruction quality, variability, and visual quality. We evaluate SEAN on multiple datasets and report better quantitative metrics (e.g. FID, PSNR) than the current state of the art. SEAN also pushes the frontier of interactive image editing. We can interactively edit images by changing segmentation masks or the style for any given region. We can also interpolate styles from two reference images per region.

2004.08247

Real-time sparse-sampled Ptychographic imaging through deep neural networks

Ptychography has rapidly grown in the fields of X-ray and electron imaging for its unprecedented ability to achieve nano or atomic scale resolution while simultaneously retrieving chemical or magnetic information from a sample. A ptychographic reconstruction is achieved by means of solving a complex inverse problem that imposes constraints both on the acquisition and on the analysis of the data, which typically precludes real-time imaging due to computational cost involved in solving this inverse problem. In this work we propose PtychoNN, a novel approach to solve the ptychography reconstruction problem based on deep convolutional neural networks. We demonstrate how the proposed method can be used to predict real-space structure and phase at each scan point solely from the corresponding far-field diffraction data. The presented results demonstrate how PtychoNN can effectively be used on experimental data, being able to generate high quality reconstructions of a sample up to hundreds of times faster than state-of-the-art ptychography reconstruction solutions once trained. By surpassing the typical constraints of iterative model-based methods, we can significantly relax the data acquisition sampling conditions and produce equally satisfactory reconstructions. Besides drastically accelerating acquisition and analysis, this capability can enable new imaging scenarios that were not possible before, in cases of dose sensitive, dynamic and extremely voluminous samples.

Deep neural networks (DNNs) have become increasingly important due to their excellent empirical performance on a wide range of problems. However, regularization is generally achieved by indirect means, largely due to the complex set of functions defined by a network and the difficulty in measuring function complexity. There exists no method in the literature for additive regularization based on a norm of the function, as is classically considered in statistical learning theory. In this work, we propose sampling-based approximations to weighted function norms as regularizers for deep neural networks. We provide, to the best of our knowledge, the first proof in the literature of the NP-hardness of computing function norms of DNNs, motivating the necessity of an approximate approach. We then derive a generalization bound for functions trained with weighted norms and prove that a natural stochastic optimization strategy minimizes the bound. Finally, we empirically validate the improved performance of the proposed regularization strategies for both convex function sets as well as DNNs on real-world classification and image segmentation tasks demonstrating improved performance over weight decay, dropout, and batch normalization. Source code will be released at the time of publication.

2004.08248

From Speech to Recital -- A case of phase transition? A non-linear study

A question has been whirling in the musical fraternity for generations. Which comes first? Melody or lyrics? The answer to which is still unknown. It is well established that the lyrics play a very important role in portraying the emotional content of a particular recitation or song. The audience intuitively listens to the lyrics of the song or recitation and tends to get an idea of the mood of the song or recitation. But what would happen if the lyric part is separated from the melody and conveyed to the audience as a separate entity altogether. Would the emotional content of the recitation remain the same or the meaning would change altogether? A recitation is a way of combining the words together so that they have a sense of rhythm and thus an emotional content is imbibed within. In this study we envisaged to answer these questions in a scientific manner taking into consideration five well known Bengali recitations of different poets conveying a variety of moods ranging from joy to sorrow. The clips were recited as well as read by the same person to avoid any perceptual difference arising out of timbre variation. Next, the emotional content from the 5 recitations were standardized with the help of listening test conducted on a pool of 50 participants. The recitations as well as the speech were analyzed with the help of a latest non linear technique called Detrended Fluctuation Analysis that gives a scaling exponent, which is essentially the measure of long range correlations present in the signal. Similar pieces were extracted from the complete signal and analyzed with the help of DFA technique. Our analysis shows that the scaling exponent for all parts of recitation were much higher in general as compared to their counterparts in speech.

This paper is devoted to a continuous Cucker-Smale model with noise, which has isotropic and polarized stationary solutions depending on the intensity of the noise. The first result establishes the threshold value of the noise parameter which drives the phase transition. This threshold value is used to classify all stationary solutions and their linear stability properties. Using an entropy, these stability properties are extended to the non-linear regime. The second result is concerned with the asymptotic behaviour of the solutions of the evolution problem. In several cases, we prove that stable solutions attract the other solutions with an optimal exponential rate of convergence determined by the spectral gap of the linearized problem around the stable solutions. The spectral gap has to be computed in a norm adapted to the non-local term.

2004.08249

Understanding the Difficulty of Training Transformers

Transformers have been proved effective for many deep learning tasks. Training transformers, however, requires non-trivial efforts regarding carefully designing learning rate schedulers and cutting-edge optimizers (the standard SGD fails to train Transformers effectively). In this paper, we study Transformer training from both theoretical and empirical perspectives. Our analysis reveals that unbalanced gradients are not the root cause of the instability of training. Instead, we identify an amplification effect that substantially influences training. Specifically, we observe that for each layer in a multi-layer Transformer model, heavy dependency on its residual branch makes training unstable since it amplifies small parameter perturbations (e.g., parameter updates) and result in significant disturbances in the model output, yet a light dependency limits the potential of model training and can lead to an inferior trained model. Inspired by our analysis, we propose Admin ($\mathbf{Ad}$aptive $\mathbf{m}$odel $\mathbf{in}$itialization) to stabilize the training in the early stage and unleash its full potential in the late stage. Extensive experiments show that Admin is more stable, converges faster, and leads to better performance.

The learning of Transformation-Equivariant Representations (TERs), which is introduced by Hinton et al. \cite{hinton2011transforming}, has been considered as a principle to reveal visual structures under various transformations. It contains the celebrated Convolutional Neural Networks (CNNs) as a special case that only equivary to the translations. In contrast, we seek to train TERs for a generic class of transformations and train them in an {\em unsupervised} fashion. To this end, we present a novel principled method by Autoencoding Variational Transformations (AVT), compared with the conventional approach to autoencoding data. Formally, given transformed images, the AVT seeks to train the networks by maximizing the mutual information between the transformations and representations. This ensures the resultant TERs of individual images contain the {\em intrinsic} information about their visual structures that would equivary {\em extricably} under various transformations in a generalized {\em nonlinear} case. Technically, we show that the resultant optimization problem can be efficiently solved by maximizing a variational lower-bound of the mutual information. This variational approach introduces a transformation decoder to approximate the intractable posterior of transformations, resulting in an autoencoding architecture with a pair of the representation encoder and the transformation decoder. Experiments demonstrate the proposed AVT model sets a new record for the performances on unsupervised tasks, greatly closing the performance gap to the supervised models.

2004.08250

How to Teach DNNs to Pay Attention to the Visual Modality in Speech Recognition

Audio-Visual Speech Recognition (AVSR) seeks to model, and thereby exploit, the dynamic relationship between a human voice and the corresponding mouth movements. A recently proposed multimodal fusion strategy, AV Align, based on state-of-the-art sequence to sequence neural networks, attempts to model this relationship by explicitly aligning the acoustic and visual representations of speech. This study investigates the inner workings of AV Align and visualises the audio-visual alignment patterns. Our experiments are performed on two of the largest publicly available AVSR datasets, TCD-TIMIT and LRS2. We find that AV Align learns to align acoustic and visual representations of speech at the frame level on TCD-TIMIT in a generally monotonic pattern. We also determine the cause of initially seeing no improvement over audio-only speech recognition on the more challenging LRS2. We propose a regularisation method which involves predicting lip-related Action Units from visual representations. Our regularisation method leads to better exploitation of the visual modality, with performance improvements between 7% and 30% depending on the noise level. Furthermore, we show that the alternative Watch, Listen, Attend, and Spell network is affected by the same problem as AV Align, and that our proposed approach can effectively help it learn visual representations. Our findings validate the suitability of the regularisation method to AVSR and encourage researchers to rethink the multimodal convergence problem when having one dominant modality.

Self-attention network, an attention-based feedforward neural network, has recently shown the potential to replace recurrent neural networks (RNNs) in a variety of NLP tasks. However, it is not clear if the self-attention network could be a good alternative of RNNs in automatic speech recognition (ASR), which processes the longer speech sequences and may have online recognition requirements. In this paper, we present a RNN-free end-to-end model: self-attention aligner (SAA), which applies the self-attention networks to a simplified recurrent neural aligner (RNA) framework. We also propose a chunk-hopping mechanism, which enables the SAA model to encode on segmented frame chunks one after another to support online recognition. Experiments on two Mandarin ASR datasets show the replacement of RNNs by the self-attention networks yields a 8.4%-10.2% relative character error rate (CER) reduction. In addition, the chunk-hopping mechanism allows the SAA to have only a 2.5% relative CER degradation with a 320ms latency. After jointly training with a self-attention network language model, our SAA model obtains further error rate reduction on multiple datasets. Especially, it achieves 24.12% CER on the Mandarin ASR benchmark (HKUST), exceeding the best end-to-end model by over 2% absolute CER.

2004.08251

Sufficient and necessary conditions for hereditary of infinite category algebras

We describe necessary and sufficient conditions for the hereditarity of the category algebra of an infinite EI category satisfying certain combinatorial assumptions. More generally, we discuss conditions such that the left global dimension of a category algebra equals the maximal left global dimension of the endomorphism algebras of its objects, and classify its projective modules in this case. As applications, we completely classify transporter categories, orbit categories, and Quillen categories with left hereditary category algebras over a field.

We study the transverse Lusternik-Schnirelmann category of a Riemannian foliation on a compact manifold. We obtain a necessary and sufficient condition when the transverse LS category is finite. We also introduce a variation on the concept of transverse LS category, the essential transverse category, and show that this is finite for every Riemannian foliation and coincides with the transverse category if the latter is finite. Moreover we prove that the essential transverse category is a lower bound for the number of critical leaf closures of a basic differentiable function on M.

2004.08252

Comparison between the non-crossing and the non-crossing on lines properties

In the recent paper [2], it was proved that the closure of the planar diffeomorphisms in the Sobolev norm consists of the functions which are non-crossing (NC), i.e., the functions which can be uniformly approximated by continuous one-to-one functions on the grids. A deep simplification of this property is to consider curves instead of grids, so considering functions which are non-crossing on lines (NCL). Since the NCL property is way easier to check, it would be extremely positive if they actually coincide, while it is only obvious that NC implies NCL. We show that in general NCL does not imply NC, but the implication becomes true with the additional assumption that $\det(Du)>0$ a.e., which is a very common assumption in nonlinear elasticity.

We calculate the Thomson scattering cross section in a non-relativistic, magnetized, high density plasma -- in a regime where collective excitations can be described by magnetohydrodynamics. We show that, in addition to cyclotron resonances and an elastic peak, the cross section exhibits two pairs of peaks associated with slow and fast magnetosonic waves; by contrast, the cross section arising in pure hydrodynamics possesses just a single pair of Brillouin peaks. Both the position and the width of these magnetosonic-wave peaks depend on the ambient magnetic field and temperature, as well as transport and thermodynamic coefficients, and so can therefore serve as a diagnostic tool for plasma properties that are otherwise challenging to measure.

2004.08253

Hadron and lepton tensors in semileptonic decays including new physics

We extend our general framework for semileptonic decay, originally introduced in N. Penalva et al. [Phys. Rev. D100, 113007 (2019)], with the addition of new physics (NP) tensor terms. In this way, all the NP effective Hamiltonians that are considered in lepton flavour universality violation (LFUV) studies have now been included. Besides, we now also give general expressions that allow for complex Wilson coefficients. The scheme developed is totally general and it can be applied to any charged current semileptonic decay, involving any quark flavors or initial and final hadron states. We show that all the hadronic input, including NP effects, can be parametrized in terms of 16 Lorentz scalar structure functions. In the second part of this work, we use this formalism to obtain the complete NP effects in the $\Lambda_b\to \Lambda_c \tau\bar\nu_\tau$ semileptonic decay, where LFUV, if finally confirmed, is also expected to be seen. We stress the relevance of the center of mass (CM) $d^2\Gamma/(d\omega d\cos\theta_\ell)$ and laboratory (LAB) $d^2\Gamma/(d\omega dE_\ell)$ differential decay widths. While models with very different strengths in the NP terms give the same differential $d\Gamma/d\omega$ and total decay widths for this decay, they predict very different numerical results for some of the $\cos\theta_\ell$ and $E_\ell$ coefficient-functions that determine the above two distributions. Thus, the combined analysis of the CM $d^2\Gamma/(d\omega d\cos\theta_\ell)$ and LAB $d^2\Gamma/ (d\omega dE_\ell)$ differential decay widths will help clarifying what kind of NP is a better candidate in order to explain LFUV.

KLOE has measured most decay branching ratios of Ks, Kl and K+/- mesons. It has also measured the Kl and the K+- lifetime and determined the shape of the form factors involved in kaon semileptonic decays. We present in the following a description of the above measurements and a well organized compendium of all of our data, with particular attention to correlations. These data provide the basis for the determination of the CKM parameter Vus and a test of the unitarity of the quark flavor mixing matrix. We also test lepton universality and place bounds on new physics using measurements of Vus from Kl2 and Kl3 decays.

2004.08254

Event reconstruction for KM3NeT/ORCA using convolutional neural networks

The KM3NeT research infrastructure is currently under construction at two locations in the Mediterranean Sea. The KM3NeT/ORCA water-Cherenkov neutrino detector off the French coast will instrument several megatons of seawater with photosensors. Its main objective is the determination of the neutrino mass ordering. This work aims at demonstrating the general applicability of deep convolutional neural networks to neutrino telescopes, using simulated datasets for the KM3NeT/ORCA detector as an example. To this end, the networks are employed to achieve reconstruction and classification tasks that constitute an alternative to the analysis pipeline presented for KM3NeT/ORCA in the KM3NeT Letter of Intent. They are used to infer event reconstruction estimates for the energy, the direction, and the interaction point of incident neutrinos. The spatial distribution of Cherenkov light generated by charged particles induced in neutrino interactions is classified as shower- or track-like, and the main background processes associated with the detection of atmospheric neutrinos are recognized. Performance comparisons to machine-learning classification and maximum-likelihood reconstruction algorithms previously developed for KM3NeT/ORCA are provided. It is shown that this application of deep convolutional neural networks to simulated datasets for a large-volume neutrino telescope yields competitive reconstruction results and performance improvements with respect to classical approaches.

Convolutional neural networks (CNNs) have achieved breakthrough performances in a wide range of applications including image classification, semantic segmentation, and object detection. Previous research on characterizing the generalization ability of neural networks mostly focuses on fully connected neural networks (FNNs), regarding CNNs as a special case of FNNs without taking into account the special structure of convolutional layers. In this work, we propose a tighter generalization bound for CNNs by exploiting the sparse and permutation structure of its weight matrices. As the generalization bound relies on the spectral norm of weight matrices, we further study spectral norms of three commonly used convolution operations including standard convolution, depthwise convolution, and pointwise convolution. Theoretical and experimental results both demonstrate that our bounds for CNNs are tighter than existing bounds.

2004.08255

A Survey of Approximate Quantile Computation on Large-scale Data (Technical Report)

As data volume grows extensively, data profiling helps to extract metadata of large-scale data. However, one kind of metadata, order statistics, is difficult to be computed because they are not mergeable or incremental. Thus, the limitation of time and memory space does not support their computation on large-scale data. In this paper, we focus on an order statistic, quantiles, and present a comprehensive analysis of studies on approximate quantile computation. Both deterministic algorithms and randomized algorithms that compute approximate quantiles over streaming models or distributed models are covered. Then, multiple techniques for improving the efficiency and performance of approximate quantile algorithms in various scenarios, such as skewed data and high-speed data streams, are presented. Finally, we conclude with coverage of existing packages in different languages and with a brief discussion of the future direction in this area.

The g-and-k and (generalised) g-and-h distributions are flexible univariate distributions which can model highly skewed or heavy tailed data through only four parameters: location and scale, and two shape parameters influencing the skewness and kurtosis. These distributions have the unusual property that they are defined through their quantile function (inverse cumulative distribution function) and their density is unavailable in closed form, which makes parameter inference complicated. This paper presents the gk R package to work with these distributions. It provides the usual distribution functions and several algorithms for inference of independent identically distributed data, including the finite difference stochastic approximation method, which has not been used before for this problem.

2004.08256

On the usage of randomized p-values in the Schweder-Spjotvoll estimator

We are concerned with multiple test problems with composite null hypotheses and the estimation of the proportion $\pi_{0}$ of true null hypotheses. The Schweder-Spj\o tvoll estimator $\hat{\pi}_0$ utilizes marginal $p$-values and only works properly if the $p$-values that correspond to the true null hypotheses are uniformly distributed on $[0,1]$ ($\mathrm{Uni}[0,1]$-distributed). In the case of composite null hypotheses, marginal $p$-values are usually computed under least favorable parameter configurations (LFCs). Thus, they are stochastically larger than $\mathrm{Uni}[0,1]$ under non-LFCs in the null hypotheses. When using these LFC-based $p$-values, $\hat{\pi}_0$ tends to overestimate $\pi_{0}$. We introduce a new way of randomizing $p$-values that depends on a tuning parameter $c\in[0,1]$, such that $c=0$ and $c=1$ lead to $\mathrm{Uni}[0,1]$-distributed $p$-values, which are independent of the data, and to the original LFC-based $p$-values, respectively. For a certain value $c=c^{\star}$ the bias of $\hat{\pi}_0$ is minimized when using our randomized $p$-values. This often also entails a smaller mean squared error of the estimator as compared to the usage of the LFC-based $p$-values. We analyze these points theoretically, and we demonstrate them numerically in computer simulations under various standard statistical models.

Extending R. A. Fisher and D. A. Freedman's results on the analysis of covariance, Lin [2013] proposed an ordinary least squares adjusted estimator of the average treatment effect in completely randomized experiments. We further study its statistical properties under the potential outcomes model in the asymptotic regimes allowing for a diverging number of covariates. We show that Lin [2013]'s estimator is consistent when $\kappa \log p \rightarrow 0$ and asymptotically normal when $\kappa p \rightarrow 0$ under mild moment conditions, where $\kappa$ is the maximum leverage score of the covariate matrix. In the favorable case where leverage scores are all close together, his estimator is consistent when $p = o(n / \log n)$ and is asymptotically normal when $p = o(n^{1/2})$. In addition, we propose a bias-corrected estimator that is consistent when $\kappa \log p\rightarrow 0$ and is asymptotically normal, with the same variance in the fixed-$p$ regime, when $\kappa^{2} p \log p \rightarrow 0$. In the favorable case, the latter condition reduces to $p = o(n^{2/3} / (\log n)^{1/3})$. Similar to Lin [2013], our results hold for non-random potential outcomes and covariates without any model specification. Our analysis requires novel analytic tools for sampling without replacement, which complement and potentially enrich the theory in other areas such as survey sampling, matrix sketching, and transductive learning.

2004.08257

Duplication Detection in Knowledge Graphs: Literature and Tools

In recent years, an increasing amount of knowledge graphs (KGs) have been created as a means to store cross-domain knowledge and billion of facts, which are the basis of costumers' applications like search engines. However, KGs inevitably have inconsistencies such as duplicates that might generate conflicting property values. Duplication detection (DD) aims to identify duplicated entities and resolve their conflicting property values effectively and efficiently. In this paper, we perform a literature review on DD methods and tools, and an evaluation of them. Our main contributions are a performance evaluation of DD tools in KGs, improvement suggestions, and a DD workflow to support future development of DD tools, which are based on desirable features detected through this study.

Representing entities and relations in an embedding space is a well-studied approach for machine learning on relational data. Existing approaches, however, primarily focus on improving accuracy and overlook other aspects such as robustness and interpretability. In this paper, we propose adversarial modifications for link prediction models: identifying the fact to add into or remove from the knowledge graph that changes the prediction for a target fact after the model is retrained. Using these single modifications of the graph, we identify the most influential fact for a predicted link and evaluate the sensitivity of the model to the addition of fake facts. We introduce an efficient approach to estimate the effect of such modifications by approximating the change in the embeddings when the knowledge graph changes. To avoid the combinatorial search over all possible facts, we train a network to decode embeddings to their corresponding graph components, allowing the use of gradient-based optimization to identify the adversarial modification. We use these techniques to evaluate the robustness of link prediction models (by measuring sensitivity to additional facts), study interpretability through the facts most responsible for predictions (by identifying the most influential neighbors), and detect incorrect facts in the knowledge base.

2004.08258

Initial forms and a notion of basis for tropical differential equations

We show that solution sets of systems of tropical differential equations can be characterised in terms of monomial-freeness of an initial ideal. We discuss a candidate definition of tropical differential basis and give a nonexistence result for such bases in an example.

Tropical and idempotent analysis with their relations to the Hamilton-Jacobi and matrix Bellman equations are discussed. Some dequantization procedures are important in tropical and idempotent mathematics. In particular, the Hamilton-Jacobi-Bellman equation is treated as a result of the Maslov dequantization applied to the Schr\"{o}dinger equation. This leads to a linearity of the Hamilton-Jacobi-Bellman equation over tropical algebras. The correspondence principle and the superposition principle of idempotent mathematics are formulated and examined. The matrix Bellman equation and its applications to optimization problems on graphs are discussed. Universal algorithms for numerical algorithms in idempotent mathematics are investigated. In particular, an idempotent version of interval analysis is briefly discussed.

2004.08259

Constrained Levenberg-Marquardt Method with Global Complexity Bound

A new Levenberg--Marquardt (LM) method for solving nonlinear least squares problems with convex constraints is described. Various versions of this method have been proposed, their main differences being in the choice of damping parameter. In this paper, we propose a new rule for updating the damping parameter that is based on the perspective of majorization-minimization method. Our method finds an $\epsilon$-stationary point within $O(\epsilon^{-2})$ iterations and converges locally quadratically under a local error bound condition. These properties are guaranteed even when the subproblems are solved inexactly, which reduces the computational cost per iteration of our method. To the best of our knowledge, this is the first LM method that has an iteration complexity bound for constrained problems. Numerical results show that our method achieves better global performance than those of two existing LM methods and the projected gradient method.

This paper considers global optimization with a black-box unknown objective function that can be non-convex and non-differentiable. Such a difficult optimization problem arises in many real-world applications, such as parameter tuning in machine learning, engineering design problem, and planning with a complex physics simulator. This paper proposes a new global optimization algorithm, called Locally Oriented Global Optimization (LOGO), to aim for both fast convergence in practice and finite-time error bound in theory. The advantage and usage of the new algorithm are illustrated via theoretical analysis and an experiment conducted with 11 benchmark test functions. Further, we modify the LOGO algorithm to specifically solve a planning problem via policy search with continuous state/action space and long time horizon while maintaining its finite-time error bound. We apply the proposed planning method to accident management of a nuclear power plant. The result of the application study demonstrates the practical utility of our method.

2004.08260

Forecasting Multi-Dimensional Processes over Graphs

The forecasting of multi-variate time processes through graph-based techniques has recently been addressed under the graph signal processing framework. However, problems in the representation and the processing arise when each time series carries a vector of quantities rather than a scalar one. To tackle this issue, we devise a new framework and propose new methodologies based on the graph vector autoregressive model. More explicitly, we leverage product graphs to model the high-dimensional graph data and develop multi-dimensional graph-based vector autoregressive models to forecast future trends with a number of parameters that is independent of the number of time series and a linear computational complexity. Numerical results demonstrating the prediction of moving point clouds corroborate our findings.

Short-term demand forecasting models commonly combine convolutional and recurrent layers to extract complex spatiotemporal patterns in data. Long-term histories are also used to consider periodicity and seasonality patterns as time series data. In this study, we propose an efficient architecture, Temporal-Guided Network (TGNet), which utilizes graph networks and temporal-guided embedding. Graph networks extract invariant features to permutations of adjacent regions instead of convolutional layers. Temporal-guided embedding explicitly learns temporal contexts from training data and is substituted for the input of long-term histories from days/weeks ago. TGNet learns an autoregressive model, conditioned on temporal contexts of forecasting targets from temporal-guided embedding. Finally, our model achieves competitive performances with other baselines on three spatiotemporal demand dataset from real-world, but the number of trainable parameters is about 20 times smaller than a state-of-the-art baseline. We also show that temporal-guided embedding learns temporal contexts as intended and TGNet has robust forecasting performances even to atypical event situations.

2004.08261

The Bogomolov-Beauville-Yau Decomposition for Klt Projective Varieties with Trivial First Chern Class -- Without Tears

We give a simplified proof (in characteristic zero) of the decomposition theorem for complex projective varieties with klt singularities and numerically trivial canonical bundle. The proof rests in an essential way on most of the partial results of the previous proof obtained by many authors, but avoids those in positive characteristic by S. Druel. The single to some extent new contribution is an algebraicity and bimeromorphic splitting result for generically locally trivial fibrations with fibres without holomorphic vector fields. We give first the proof in the easier smooth case, following the same steps as in the general case, treated next.

We generalize the Chern class relation for the transversal intersection of two nonsingular varieties to a relation for possibly singular varieties, under a 'splayedness' assumption. The relation is shown to hold for both the Chern-Schwartz-MacPherson class and the Chern-Fulton class. The main tool is a formula for Segre classes of splayed subschemes. We also discuss the Chern class relation under the assumption that one of the varieties is a general very ample divisor.

2004.08262

Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques

Machine-learning (ML) techniques are explored to identify and classify hadronic decays of highly Lorentz-boosted W/Z/Higgs bosons and top quarks. Techniques without ML have also been evaluated and are included for comparison. The identification performances of a variety of algorithms are characterized in simulated events and directly compared with data. The algorithms are validated using proton-proton collision data at $\sqrt{s} =$ 13 TeV, corresponding to an integrated luminosity of 35.9 fb$^{-1}$. Systematic uncertainties are assessed by comparing the results obtained using simulation and collision data. The new techniques studied in this paper provide significant performance improvements over non-ML techniques, reducing the background rate by up to an order of magnitude at the same signal efficiency.

Since the machine learning techniques are improving rapidly, it has been shown that the image recognition techniques in deep neural networks can be used to detect jet substructure. And it turns out that deep neural networks can match or outperform traditional approach of expert features. However, there are disadvantages such as sparseness of jet images. Based on the natural tree-like structure of jet sequential clustering, the recursive neural networks (RecNNs), which embed jet clustering history recursively as in natural language processing, have a better behavior when confronted with these problems. We thus try to explore the performance of RecNNs in quark/gluon discrimination. The results show that RecNNs work better than the baseline boosted decision tree (BDT) by a few percent in gluon rejection rate. However, extra implementation of particle flow identification only increases the performance slightly. We also experimented on some relevant aspects which might influence the performance of the networks. It shows that even taking only particle flow identification as input feature without any extra information on momentum or angular position is already giving a fairly good result, which indicates that the most of the information for quark/gluon discrimination is already included in the tree-structure itself. As a bonus, a rough up/down quark jets discrimination is also explored.

2004.08263

Leveraging Mobility Flows from Location Technology Platforms to Test Crime Pattern Theory in Large Cities

Crime has been previously explained by social characteristics of the residential population and, as stipulated by crime pattern theory, might also be linked to human movements of non-residential visitors. Yet a full empirical validation of the latter is lacking. The prime reason is that prior studies are limited to aggregated statistics of human visitors rather than mobility flows and, because of that, neglect the temporal dynamics of individual human movements. As a remedy, we provide the first work which studies the ability of granular human mobility in describing and predicting crime concentrations at an hourly scale. For this purpose, we propose the use of data from location technology platforms. This type of data allows us to trace individual transitions and, therefore, we succeed in distinguishing different mobility flows that (i) are incoming or outgoing from a neighborhood, (ii) remain within it, or (iii) refer to transitions where people only pass through the neighborhood. Our evaluation infers mobility flows by leveraging an anonymized dataset from Foursquare that includes almost 14.8 million consecutive check-ins in three major U.S. cities. According to our empirical results, mobility flows are significantly and positively linked to crime. These findings advance our theoretical understanding, as they provide confirmatory evidence for crime pattern theory. Furthermore, our novel use of digital location services data proves to be an effective tool for crime forecasting. It also offers unprecedented granularity when studying the connection between human mobility and crime.

With the wide adoption of the multi-community setting in many popular social media platforms, the increasing user engagements across multiple online communities warrant research attention. In this paper, we introduce a novel analogy between the movements in the cyber space and the physical space. This analogy implies a new way of studying human online activities by modelling the activities across online communities in a similar fashion as the movements among locations. First, we quantitatively validate the analogy by comparing several important properties of human online activities and physical movements. Our experiments reveal striking similarities between the cyber space and the physical space. Next, inspired by the established methodology on human mobility in the physical space, we propose a framework to study human "mobility" across online platforms. We discover three interesting patterns of user engagements in online communities. Furthermore, our experiments indicate that people with different mobility patterns also exhibit divergent preferences to online communities. This work not only attempts to achieve a better understanding of human online activities, but also intends to open a promising research direction with rich implications and applications.

2004.08264

Machine learning reveals complex behaviours in optically trapped particles

Since their invention in the 1980s [1], optical tweezers have found a wide range of applications, from biophotonics and mechanobiology to microscopy and optomechanics [2, 3, 4, 5]. Simulations of the motion of microscopic particles held by optical tweezers are often required to explore complex phenomena and to interpret experimental data [6, 7, 8, 9]. For the sake of computational efficiency, these simulations usually model the optical tweezers as an harmonic potential [10, 11]. However, more physically-accurate optical-scattering models [12, 13, 14, 15] are required to accurately model more onerous systems; this is especially true for optical traps generated with complex fields [16, 17, 18, 19]. Although accurate, these models tend to be prohibitively slow for problems with more than one or two degrees of freedom (DoF) [20], which has limited their broad adoption. Here, we demonstrate that machine learning permits one to combine the speed of the harmonic model with the accuracy of optical-scattering models. Specifically, we show that a neural network can be trained to rapidly and accurately predict the optical forces acting on a microscopic particle. We demonstrate the utility of this approach on two phenomena that are prohibitively slow to accurately simulate otherwise: the escape dynamics of swelling microparticles in an optical trap, and the rotation rates of particles in a superposition of beams with opposite orbital angular momenta. Thanks to its high speed and accuracy, this method can greatly enhance the range of phenomena that can be efficiently simulated and studied.

Particle size measurement based on digital holography with conventional algorithms are usually time-consuming and susceptible to noises associated with hologram quality and particle complexity, limiting its usage in a broad range of engineering applications and fundamental research. We propose a learning-based hologram processing method to cope with the aforementioned issues. The proposed approach uses a modified U-net architecture with three input channels and two output channels, and specially-designed loss functions. The proposed method has been assessed using synthetic, manually-labeled experimental, and water tunnel bubbly flow data containing particles of different shapes. The results demonstrate that our approach can achieve better performance in comparison to the state-of-the-art non-machine-learning methods in terms of particle extraction rate and positioning accuracy with significantly improved processing speed. Our learning-based approach can be extended to other types of image-based particle size measurements.

2004.08265

Two-Pion production in the second resonance region in $\pi^- p$ collisions with HADES

Pion induced reactions provide unique opportunities for an unambiguous description of baryonic resonances and their coupling channels by means of a partial wave analysis. Using the secondary pion beam at SIS18, the two pion production in the second resonance region has been investigated to unravel the role of the $N(1520) \frac{3}{2}^-$ resonance in the intermediate $\rho$ production. Results on exclusive channels with one pion ($\pi^{-}p$) and two pions ($\pi^{+}\pi^{-}n$, $\pi^{0}\pi^{-}p$) in the final state measured in the $\pi^{-}-p$ reaction at four different pion beam momenta (0.650, 0.685, 0.733, and 0.786 GeV/c) are presented. The excitation function of the different partial waves and $\Delta\pi$, $N\sigma$ and $N\rho$ isobar configurations is obtained, using the Bonn-Gatchina partial wave analysis. The $N(1520) \frac{3}{2}^-$ resonance is found to dominate the $N\rho$ final state with the branching ratio $BR=12.2 \pm 1.9\%$.

We consider hadron production in deuteron gold collisions at RHIC in the forward rapidity region. Treating the target nucleus as a Color Glass Condensate and the projectile deuteron as a dilute system of valence quarks, we obtain good agreement with the BRAHMS minimum bias data on charged hadron production in the forward rapidity ($y=3.2$) and low $p_t$ region. We provide predictions for neutral pion production in minimum bias deuteron gold collisions in the forward rapidity region, $y=3.8$, measured by the STAR collaboration at RHIC.

2004.08266

The 2-rank of the class group of some real cyclic quartic number fields II

In this paper, we determine the 2-rank of the class group of certain classes of real cyclic quartic number fields. Precisely, we consider the case in which the quadratic subfield is Q(\sqrt{l}) with l=2 or a prime congruent to 1 mod 8.

Let A be a regular domain of dimension d containing an infinite field and let n be an integer with 2n\geq d+3. For a stably free A-module P of rank n, we prove that (i) P has a unimodular element if and only if the euler class of P is zero in E^n(A) and (ii) we define Whitney class homomorphism w(P):E^s(A)\ra E^{n+s}(A), where E^s(A) denotes the sth Euler class group of A for s\geq 1.

2004.08267

To switch or not to switch -- a machine learning approach for ferroelectricity

With the advent of increasingly elaborate experimental techniques in physics, chemistry and materials sciences, measured data are becoming bigger and more complex. The observables are typically a function of several stimuli resulting in multidimensional data sets spanning a range of experimental parameters. As an example, a common approach to study ferroelectric switching is to observe effects of applied electric field, but switching can also be enacted by pressure and is influenced by strain fields, material composition, temperature, time, etc. Moreover, the parameters are usually interdependent, so that their decoupling toward univariate measurements or analysis may not be straightforward. On the other hand, both explicit and hidden parameters provide an opportunity to gain deeper insight into the measured properties, provided there exists a well-defined path to capture and analyze such data. Here, we introduce a new, two-dimensional approach to represent hysteretic response of a material system to applied electric field. Utilizing ferroelectric polarization as a model hysteretic property, we demonstrate how explicit consideration of electromechanical response to two rather than one control voltages enables significantly more transparent and robust interpretation of observed hysteresis, such as differentiating between charge trapping and ferroelectricity. Furthermore, we demonstrate how the new data representation readily fits into a variety of machinelearning methodologies, from unsupervised classification of the origins of hysteretic response via linear clustering algorithms to neural-network-based inference of the sample temperature based on the specific morphology of hysteresis.

The influence of depolarizing field on the magnitude and stability of a uniform polarization in ferroelectric capacitors and tunnel junctions is studied using a nonlinear thermodynamic theory. It is predicted that, in heterostructures involving strained epitaxial films and metal electrodes, the homogeneous polarization state may remain stable against transformations into the paraelectric phase and into polydomain states down to the nanoscale. This result supports the possibility of depolarizing-field-related resistive switching in ferroelectric tunnel junctions with dissimilar electrodes. The resistance on/off ratio in such junctions is shown to be governed by the difference between the reciprocal capacitances of screening space charges in the electrodes.

2004.08268

Development of gamma insensitive silicon carbide diagnostics to qualify intense thermal and epithermal neutron fields

The e_LiBANS project aims at creating accelerator based compact neutron facilities for diverse interdisciplinary applications. After the successful setting up and characterization of a thermal neutron source based on a medical electron LINAC, a similar assembly for epithermal neutrons has been developed. The project is based on an Elekta 18 MV LINAC coupled with a photoconverter-moderator system which deploys the ({\gamma},n) photonuclear reaction to convert a bremsstrahlung photon beam into a neutron field. This communication describes the development of novel diagnostics to qualify the thermal and epithermal neutron fields that have been produced. In particular, a proof of concept for the use of silicon carbide photodiodes as a thermal neutron rate detector is presented.

Silicon Carbide (SiC) is a typical material for third-generation semiconductor. The thermal boundary resistance (TBR) of 4H-SiC/SiO2 interface, was investigated by both experimental measurements and theoretical calculations. The structure of 4H-SiC/SiO2 was characterized by using transmission electron microscopy and X-ray diffraction. The TBR is measured as 8.11*10-8 m2K/W by 3-omega method. Furthermore, the diffuse mismatch model was employed to predict the TBR of different interfaces which is in good agreement with measurements. Heat transport behavior based on phonon scattering perspective was also discussed to understand the variations of TBR across different interfaces. Besides, the intrinsic thermal conductivity of SiO2 thin films (200~1,500 nm in thickness) on 4H-SiC substrates was measured by 3 omega procedure, as 1.42 W/mK at room temperature. It is believed the presented results could provide useful insights on the thermal management and heat dissipation for SiC devices.

2004.08269

Beat Detection and Automatic Annotation of the Music of Bharatanatyam Dance using Speech Recognition Techniques

Bharatanatyam, an Indian Classical Dance form, represents the rich cultural heritage of India. Analysis and recognition of such dance forms are critical for the preservation of cultural heritage. Like in most dance forms, a Bharatanatyam dancer performs in synchronization with structured rhythmic music, called Sollukattu, which comprises instrumental beats and vocalized utterances (bols) to create a rhythmic music structure. Computer analysis of Bharatanatyam, therefore, requires a structural analysis of Sollukattus. In this paper, we use speech processing techniques to recognize bols. Exploiting the predefined structures of Sollukattus and the detected bols, we recognize the Sollukattu. We estimate the tempo period by two methods. Finally, we generate a complete annotation of the audio signal by beat marking. For this, we also use the information of beats detected from the onset envelope of a Sollukattu signal. For training and test, we create a data set for Sollukattus and annotate them. We achieve 85% accuracy in bol recognition, 95% in Sollukattu recognition, 96% in tempo period estimation, and over 90% in beat marking. This is the maiden attempt to fully structurally analyze the music of an Indian Classical Dance form and the use of speech processing techniques for beat marking.

Code-switching (CS) detection refers to the automatic detection of language switches in code-mixed utterances. This task can be achieved by using a CS automatic speech recognition (ASR) system that can handle such language switches. In our previous work, we have investigated the code-switching detection performance of the Frisian-Dutch CS ASR system by using the time alignment of the most likely hypothesis and found that this technique suffers from over-switching due to numerous very short spurious language switches. In this paper, we propose a novel method for CS detection aiming to remedy this shortcoming by using the language posteriors which are the sum of the frame-level posteriors of phones belonging to the same language. The CS ASR-generated language posteriors contain more complete language-specific information on frame level compared to the time alignment of the ASR output. Hence, it is expected to yield more accurate and robust CS detection. The CS detection experiments demonstrate that the proposed language posterior-based approach provides higher detection accuracy than the baseline system in terms of equal error rate. Moreover, a detailed CS detection error analysis reveals that using language posteriors reduces the false alarms and results in more robust CS detection.

2004.08270

Weakly Supervised Geodesic Segmentation of Egyptian Mummy CT Scans

In this paper, we tackle the task of automatically analyzing 3D volumetric scans obtained from computed tomography (CT) devices. In particular, we address a particular task for which data is very limited: the segmentation of ancient Egyptian mummies CT scans. We aim at digitally unwrapping the mummy and identify different segments such as body, bandages and jewelry. The problem is complex because of the lack of annotated data for the different semantic regions to segment, thus discouraging the use of strongly supervised approaches. We, therefore, propose a weakly supervised and efficient interactive segmentation method to solve this challenging problem. After segmenting the wrapped mummy from its exterior region using histogram analysis and template matching, we first design a voxel distance measure to find an approximate solution for the body and bandage segments. Here, we use geodesic distances since voxel features as well as spatial relationship among voxels is incorporated in this measure. Next, we refine the solution using a GrabCut based segmentation together with a tracking method on the slices of the scan that assigns labels to different regions in the volume, using limited supervision in the form of scribbles drawn by the user. The efficiency of the proposed method is demonstrated using visualizations and validated through quantitative measures and qualitative unwrapping of the mummy.

The absence of large scale datasets with pixel-level supervisions is a significant obstacle for the training of deep convolutional networks for scene text segmentation. For this reason, synthetic data generation is normally employed to enlarge the training dataset. Nonetheless, synthetic data cannot reproduce the complexity and variability of natural images. In this paper, a weakly supervised learning approach is used to reduce the shift between training on real and synthetic data. Pixel-level supervisions for a text detection dataset (i.e. where only bounding-box annotations are available) are generated. In particular, the COCO-Text-Segmentation (COCO_TS) dataset, which provides pixel-level supervisions for the COCO-Text dataset, is created and released. The generated annotations are used to train a deep convolutional neural network for semantic segmentation. Experiments show that the proposed dataset can be used instead of synthetic data, allowing us to use only a fraction of the training samples and significantly improving the performances.

2004.08271

Representations of fusion categories and their commutants

A bicommutant category is a higher categorical analog of a von Neumann algebra. We study the bicommutant categories which arise as the commutant $\mathcal{C}'$ of a fully faithful representation $\mathcal{C}\to\operatorname{Bim}(R)$ of a unitary fusion category $\mathcal{C}$. Using results of Izumi, Popa, and Tomatsu about existence and uniqueness of representations of unitary (multi)fusion categories, we prove that if $\mathcal{C}$ and $\mathcal{D}$ are Morita equivalent unitary fusion categories, then their commutant categories $\mathcal{C}'$ and $\mathcal{D}'$ are equivalent as bicommutant categories. In particular, they are equivalent as tensor categories: \[ \Big(\,\,\mathcal{C}\,\,\simeq_{\text{Morita}}\,\,\mathcal{D}\,\,\Big) \qquad\Longrightarrow\qquad \Big(\,\,\mathcal{C}'\,\,\simeq_{\text{tensor}}\,\,\mathcal{D}'\,\,\Big). \] This categorifies the well-known result according to which the commutants (in some representations) of Morita equivalent finite dimensional $\rm C^*$-algebras are isomorphic von Neumann algebras, provided the representations are `big enough'. We also introduce a notion of positivity for bi-involutive tensor categories. For dagger categories, positivity is a property (the property of being a $\rm C^*$-category). But for bi-involutive tensor categories, positivity is extra structure. We show that unitary fusion categories and $\operatorname{Bim}(R)$ admit distinguished positive structures, and that fully faithful representations $\mathcal{C}\to\operatorname{Bim}(R)$ automatically respect these positive structures.

Fusion rules generalize groups by allowing multivalued multiplication. Groups are fusion rules of simple current index 1. We classify nilpotent (in the sense of Gelaki and Nikshych) fusion rules of simple current index 2, and characterize the associated fusion categories.

2004.08272

Board Games for Quantum Computers

Scalable board games, including Five in a Row (or gomoku) and weiqi (or go), are generalized so that they can be played on or by quantum computers. We adopt three principles for the generalization: the first two are to ensure that the games are compatible with quantum computer and the third is to ensure that the standard classical games are the special cases. We demonstrate how to construct basic quantum moves and use them to set up quantum games. There are three different schemes to play the quantized games: one quantum computer with another quantum computer (QwQ), two classical computer playing with each other on one quantum computer (CQC), and one classical computer with another classical computer(CwC). We illustrate these results with the games of Five in a Row and weiqi.

The paper is devoted to quantization of extensive games with the use of both the Marinatto-Weber and the Eisert-Wilkens-Lewenstein concept of quantum game. We revise the current conception of quantum ultimatum game and we show why the proposal is unacceptable. To support our comment, we present the new idea of the quantum ultimatum game. Our scheme also makes a point of departure for a protocol to quantize extensive games.

2004.08273

3-D Context Entropy Model for Improved Practical Image Compression

In this paper, we present our image compression framework designed for CLIC 2020 competition. Our method is based on Variational AutoEncoder (VAE) architecture which is strengthened with residual structures. In short, we make three noteworthy improvements here. First, we propose a 3-D context entropy model which can take advantage of known latent representation in current spatial locations for better entropy estimation. Second, a light-weighted residual structure is adopted for feature learning during entropy estimation. Finally, an effective training strategy is introduced for practical adaptation with different resolutions. Experiment results indicate our image compression method achieves 0.9775 MS-SSIM on CLIC validation set and 0.9809 MS-SSIM on test set.

Computer vision tasks are often expected to be executed on compressed images. Classical image compression standards like JPEG 2000 are widely used. However, they do not account for the specific end-task at hand. Motivated by works on recurrent neural network (RNN)-based image compression and three-dimensional (3D) reconstruction, we propose unified network architectures to solve both tasks jointly. These joint models provide image compression tailored for the specific task of 3D reconstruction. Images compressed by our proposed models, yield 3D reconstruction performance superior as compared to using JPEG 2000 compression. Our models significantly extend the range of compression rates for which 3D reconstruction is possible. We also show that this can be done highly efficiently at almost no additional cost to obtain compression on top of the computation already required for performing the 3D reconstruction task.

2004.08274

Observation of symmetry protected zero modes in topolectrical circuits

Higher-order topological insulators are a new class of topological phases of matter, originally conceived for electrons in solids. It has been suggested that $\mathbb{Z}_N$ Berry phase (Berry phase quantized into $2\pi/N$) is a useful tool to characterize the symmetry protected topological states, while the experimental evidence is still elusive. Recently, topolectrical circuits have emerged as a simple yet very powerful platform for studying topological physics that are challenging to realize in condensed matter systems. Here, we present the first experimental observation of second-order corner states characterized by $\mathbb{Z}_3$ Berry phase in topolectrical circuits. We demonstrate theoretically and experimentally that the localized second-order topological states are protected by a generalized chiral symmetry of tripartite lattices, and they are pinned to "zero energy". By introducing extra capacitors within sublattices in the circuit, we are able to examine the robustness of the zero modes against both chiral-symmetry conserving and breaking disturbances. Our work paves the way for testing exotic topological band theory by electrical-circuit experiments.

We show that a quantum phase transition can occur in a phonon system in the presence of dislocations. Due to the competing nature between the topological protection of the dislocation and anharmonicity, phonons can reach a quantum critical point at a frequency determined by dislocation density and the anharmonic constant, at zero temperature. In the symmetry-broken phase, a novel phonon state is developed with a dynamically-induced dipole field. We carry out a renormalization group analysis and show that the phonon critical behavior differs wildly from any electronic system. In particular, at the critical point, a single phonon mode dominates the density of states and develops an exotic logarithmic divergence in thermal conductivity. This phonon quantum criticality provides a completely new avenue to tailor phonon transport at the single-mode level without using phononic crystals.

2004.08275

The Bernstein problem for elliptic Weingarten multigraphs

We prove that any complete, uniformly elliptic Weingarten surface in Euclidean $3$-space whose Gauss map image omits an open hemisphere is a cylinder or a plane. This generalizes a classical theorem by Hoffman, Osserman and Schoen for constant mean curvature surfaces. In particular, this proves that planes are the only complete, uniformly elliptic Weingarten multigraphs. We also show that this result holds for a large class of non-uniformly elliptic Weingarten equations. In particular, this solves in the affirmative the Bernstein problem for entire graphs for that class of elliptic equations.

We introduce here a generalization of the modified Bernstein polynomials for Jacobi weights using the $q$-Bernstein basis proposed by G.M. Phillips to generalize classical Bernstein Polynomials. The function is evaluated at points which are in geometric progression in $]0,1[$. Numerous properties of the modified Bernstein Polynomials are extended to their $q$-analogues: simultaneous approximation, pointwise convergence even for unbounded functions, shape-preserving property, Voronovskaya theorem, self-adjointness. Some properties of the eigenvectors, which are $q$-extensions of Jacobi polynomials, are given.

2004.08276

Longitudinal spin fluctuations in bcc and liquid Fe at high temperature and pressure calculated with a supercell approach

Investigation of magnetic materials at realistic conditions with first-principles methods is a challenging task due to the interplay of vibrational and magnetic degrees of freedom. The most difficult contribution to include in simulations is represented by the longitudinal magnetic degrees of freedom (LSF) due to their inherent many-body nature; nonetheless, schemes that enable to take into account this effect on a semiclassical level have been proposed and employed in the investigation of magnetic systems. However, assessment of the effect of vibrations on LSF is lacking in the literature. For this reason, in this work we develop a supercell approach within the framework of constrained density functional theory to calculate self-consistently the size of local-environment-dependent magnetic moments in the paramagnetic, high-temperature state in presence of lattice vibrations and for liquid Fe in different conditions. First, we consider the case of bcc Fe at the Curie temperature and ambient pressure. Then, we perform a similar analysis on bcc Fe at Earth's inner core conditions, and we find that LSF stabilize non-zero moments which affect atomic forces and electronic density of states of the system. Finally, we employ the present scheme on liquid Fe at the melting point at ambient pressure, and at Earth's outer core conditions ($p \approx 200$ GPa, $T \approx 6000$ K). In both cases, we obtain local magnetic moments of sizes comparable to the solid-state counterparts.

The fundamental excitations in an antiferromagnetic chain of spins-1/2 are spinons, de-confined fractional quasiparticles that when combined in pairs, form a triplet excitation continuum. In an Ising-like spin chain the continuum is gapped and the ground state is N{\'e}el ordered. Here, we report high resolution neutron scattering experiments, which reveal how a magnetic field closes this gap and drives the spin chains in \YPP\ to a critical, disordered Luttinger-liquid state. In \YPP\ the effective spins-1/2 describe the dynamics of large, Ising-like Yb magnetic moments, ensuring that the measured excitations are exclusively longitudinal, which we find to be well described by time-dependent density matrix renormalization group calculations. The inter-chain coupling leads to the confinement of spinons, a condensed matter analog of quark confinement in quantum chromodynamics. Insensitive to transverse fluctuations, our measurements show how a gapless, dispersive longitudinal mode arises from confinement and evolves with magnetic order.