Datasets:

sg247

/

multilabel-classification

like 0

Reconstructing Subject-Specific Effect Maps

Predictive models allow subject-specific inference when analyzing disease related alterations in neuroimaging data. Given a subject's data, inference can be made at two levels: global, i.e. identifiying condition presence for the subject, and local, i.e. detecting condition effect on each individual measurement extracted from the subject's data. While global inference is widely used, local inference, which can be used to form subject-specific effect maps, is rarely used because existing models often yield noisy detections composed of dispersed isolated islands. In this article, we propose a reconstruction method, named RSM, to improve subject-specific detections of predictive modeling approaches and in particular, binary classifiers. RSM specifically aims to reduce noise due to sampling error associated with using a finite sample of examples to train classifiers. The proposed method is a wrapper-type algorithm that can be used with different binary classifiers in a diagnostic manner, i.e. without information on condition presence. Reconstruction is posed as a Maximum-A-Posteriori problem with a prior model whose parameters are estimated from training data in a classifier-specific fashion. Experimental evaluation is performed on synthetically generated data and data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Results on synthetic data demonstrate that using RSM yields higher detection accuracy compared to using models directly or with bootstrap averaging. Analyses on the ADNI dataset show that RSM can also improve correlation between subject-specific detections in cortical thickness data and non-imaging markers of Alzheimer's Disease (AD), such as the Mini Mental State Examination Score and Cerebrospinal Fluid amyloid-$\beta$ levels. Further reliability studies on the longitudinal ADNI dataset show improvement on detection reliability when RSM is used.

Rotation Invariance Neural Network

Rotation invariance and translation invariance have great values in image recognition tasks. In this paper, we bring a new architecture in convolutional neural network (CNN) named cyclic convolutional layer to achieve rotation invariance in 2-D symbol recognition. We can also get the position and orientation of the 2-D symbol by the network to achieve detection purpose for multiple non-overlap target. Last but not least, this architecture can achieve one-shot learning in some cases using those invariance.

Spherical polyharmonics and Poisson kernels for polyharmonic functions

We introduce and develop the notion of spherical polyharmonics, which are a natural generalisation of spherical harmonics. In particular we study the theory of zonal polyharmonics, which allows us, analogously to zonal harmonics, to construct Poisson kernels for polyharmonic functions on the union of rotated balls. We find the representation of Poisson kernels and zonal polyharmonics in terms of the Gegenbauer polynomials. We show the connection between the classical Poisson kernel for harmonic functions on the ball, Poisson kernels for polyharmonic functions on the union of rotated balls, and the Cauchy-Hua kernel for holomorphic functions on the Lie ball.

A finite element approximation for the stochastic Maxwell--Landau--Lifshitz--Gilbert system

The stochastic Landau--Lifshitz--Gilbert (LLG) equation coupled with the Maxwell equations (the so called stochastic MLLG system) describes the creation of domain walls and vortices (fundamental objects for the novel nanostructured magnetic memories). We first reformulate the stochastic LLG equation into an equation with time-differentiable solutions. We then propose a convergent $\theta$-linear scheme to approximate the solutions of the reformulated system. As a consequence, we prove convergence of the approximate solutions, with no or minor conditions on time and space steps (depending on the value of $\theta$). Hence, we prove the existence of weak martingale solutions of the stochastic MLLG system. Numerical results are presented to show applicability of the method.

Comparative study of Discrete Wavelet Transforms and Wavelet Tensor Train decomposition to feature extraction of FTIR data of medicinal plants

Fourier-transform infra-red (FTIR) spectra of samples from 7 plant species were used to explore the influence of preprocessing and feature extraction on efficiency of machine learning algorithms. Wavelet Tensor Train (WTT) and Discrete Wavelet Transforms (DWT) were compared as feature extraction techniques for FTIR data of medicinal plants. Various combinations of signal processing steps showed different behavior when applied to classification and clustering tasks. Best results for WTT and DWT found through grid search were similar, significantly improving quality of clustering as well as classification accuracy for tuned logistic regression in comparison to original spectra. Unlike DWT, WTT has only one parameter to be tuned (rank), making it a more versatile and easier to use as a data processing tool in various signal processing applications.

On maximizing the fundamental frequency of the complement of an obstacle

Let $\Omega \subset \mathbb{R}^n$ be a bounded domain satisfying a Hayman-type asymmetry condition, and let $ D $ be an arbitrary bounded domain referred to as "obstacle". We are interested in the behaviour of the first Dirichlet eigenvalue $ \lambda_1(\Omega \setminus (x+D)) $. First, we prove an upper bound on $ \lambda_1(\Omega \setminus (x+D)) $ in terms of the distance of the set $ x+D $ to the set of maximum points $ x_0 $ of the first Dirichlet ground state $ \phi_{\lambda_1} > 0 $ of $ \Omega $. In short, a direct corollary is that if \begin{equation} \mu_\Omega := \max_{x}\lambda_1(\Omega \setminus (x+D)) \end{equation} is large enough in terms of $ \lambda_1(\Omega) $, then all maximizer sets $ x+D $ of $ \mu_\Omega $ are close to each maximum point $ x_0 $ of $ \phi_{\lambda_1} $. Second, we discuss the distribution of $ \phi_{\lambda_1(\Omega)} $ and the possibility to inscribe wavelength balls at a given point in $ \Omega $. Finally, we specify our observations to convex obstacles $ D $ and show that if $ \mu_\Omega $ is sufficiently large with respect to $ \lambda_1(\Omega) $, then all maximizers $ x+D $ of $ \mu_\Omega $ contain all maximum points $ x_0 $ of $ \phi_{\lambda_1(\Omega)} $.

On the rotation period and shape of the hyperbolic asteroid 1I/`Oumuamua (2017) U1 from its lightcurve

We observed the newly discovered hyperbolic minor planet 1I/`Oumuamua (2017 U1) on 2017 October 30 with Lowell Observatory's 4.3-m Discovery Channel Telescope. From these observations, we derived a partial lightcurve with peak-to-trough amplitude of at least 1.2 mag. This lightcurve segment rules out rotation periods less than 3 hr and suggests that the period is at least 5 hr. On the assumption that the variability is due to a changing cross section, the axial ratio is at least 3:1. We saw no evidence for a coma or tail in either individual images or in a stacked image having an equivalent exposure time of 9000 s.

Adverse effects of polymer coating on heat transport at solid-liquid interface

The ability of metallic nanoparticles to supply heat to a liquid environment under exposure to an external optical field has attracted growing interest for biomedical applications. Controlling the thermal transport properties at a solid-liquid interface then appears to be particularly relevant. In this work, we address the thermal transport between water and a gold surface coated by a polymer layer. Using molecular dynamics simulations, we demonstrate that increasing the polymer density displaces the domain resisting to the heat flow, while it doesn't affect the final amount of thermal energy released in the liquid. This unexpected behavior results from a trade-off established by the increasing polymer density which couples more efficiently with the solid but initiates a counterbalancing resistance with the liquid.

SPH calculations of Mars-scale collisions: the role of the Equation of State, material rheologies, and numerical effects

We model large-scale ($\approx$2000km) impacts on a Mars-like planet using a Smoothed Particle Hydrodynamics code. The effects of material strength and of using different Equations of State on the post-impact material and temperature distributions are investigated. The properties of the ejected material in terms of escaping and disc mass are analysed as well. We also study potential numerical effects in the context of density discontinuities and rigid body rotation. We find that in the large-scale collision regime considered here (with impact velocities of 4km/s), the effect of material strength is substantial for the post-impact distribution of the temperature and the impactor material, while the influence of the Equation of State is more subtle and present only at very high temperatures.

$\mathcal{R}_{0}$ fails to predict the outbreak potential in the presence of natural-boosting immunity

Time varying susceptibility of host at individual level due to waning and boosting immunity is known to induce rich long-term behavior of disease transmission dynamics. Meanwhile, the impact of the time varying heterogeneity of host susceptibility on the shot-term behavior of epidemics is not well-studied, even though the large amount of the available epidemiological data are the short-term epidemics. Here we constructed a parsimonious mathematical model describing the short-term transmission dynamics taking into account natural-boosting immunity by reinfection, and obtained the explicit solution for our model. We found that our system show "the delayed epidemic", the epidemic takes off after negative slope of the epidemic curve at the initial phase of epidemic, in addition to the common classification in the standard SIR model, i.e., "no epidemic" as $\mathcal{R}_{0}\leq1$ or normal epidemic as $\mathcal{R}_{0}>1$. Employing the explicit solution we derived the condition for each classification.

A global sensitivity analysis and reduced order models for hydraulically-fractured horizontal wells

We present a systematic global sensitivity analysis using the Sobol method which can be utilized to rank the variables that affect two quantity of interests -- pore pressure depletion and stress change -- around a hydraulically-fractured horizontal well based on their degree of importance. These variables include rock properties and stimulation design variables. A fully-coupled poroelastic hydraulic fracture model is used to account for pore pressure and stress changes due to production. To ease the computational cost of a simulator, we also provide reduced order models (ROMs), which can be used to replace the complex numerical model with a rather simple analytical model, for calculating the pore pressure and stresses at different locations around hydraulic fractures. The main findings of this research are: (i) mobility, production pressure, and fracture half-length are the main contributors to the changes in the quantities of interest. The percentage of the contribution of each parameter depends on the location with respect to pre-existing hydraulic fractures and the quantity of interest. (ii) As the time progresses, the effect of mobility decreases and the effect of production pressure increases. (iii) These two variables are also dominant for horizontal stresses at large distances from hydraulic fractures. (iv) At zones close to hydraulic fracture tips or inside the spacing area, other parameters such as fracture spacing and half-length are the dominant factors that affect the minimum horizontal stress. The results of this study will provide useful guidelines for the stimulation design of legacy wells and secondary operations such as refracturing and infill drilling.

Role-separating ordering in social dilemmas controlled by topological frustration

"Three is a crowd" is an old proverb that applies as much to social interactions, as it does to frustrated configurations in statistical physics models. Accordingly, social relations within a triangle deserve special attention. With this motivation, we explore the impact of topological frustration on the evolutionary dynamics of the snowdrift game on a triangular lattice. This topology provides an irreconcilable frustration, which prevents anti-coordination of competing strategies that would be needed for an optimal outcome of the game. By using different strategy updating protocols, we observe complex spatial patterns in dependence on payoff values that are reminiscent to a honeycomb-like organization, which helps to minimize the negative consequence of the topological frustration. We relate the emergence of these patterns to the microscopic dynamics of the evolutionary process, both by means of mean-field approximations and Monte Carlo simulations. For comparison, we also consider the same evolutionary dynamics on the square lattice, where of course the topological frustration is absent. However, with the deletion of diagonal links of the triangular lattice, we can gradually bridge the gap to the square lattice. Interestingly, in this case the level of cooperation in the system is a direct indicator of the level of topological frustration, thus providing a method to determine frustration levels in an arbitrary interaction network.

Dynamics of exciton magnetic polarons in CdMnSe/CdMgSe quantum wells: the effect of self-localization

We study the exciton magnetic polaron (EMP) formation in (Cd,Mn)Se/(Cd,Mg)Se diluted-magnetic-semiconductor quantum wells using time-resolved photoluminescence (PL). The magnetic field and temperature dependencies of this dynamics allow us to separate the non-magnetic and magnetic contributions to the exciton localization. We deduce the EMP energy of 14 meV, which is in agreement with time-integrated measurements based on selective excitation and the magnetic field dependence of the PL circular polarization degree. The polaron formation time of 500 ps is significantly longer than the corresponding values reported earlier. We propose that this behavior is related to strong self-localization of the EMP, accompanied with a squeezing of the heavy-hole envelope wavefunction. This conclusion is also supported by the decrease of the exciton lifetime from 600 ps to 200 - 400 ps with increasing magnetic field and temperature.

On Varieties of Ordered Automata

The classical Eilenberg correspondence, based on the concept of the syntactic monoid, relates varieties of regular languages with pseudovarieties of finite monoids. Various modifications of this correspondence appeared, with more general classes of regular languages on one hand and classes of more complex algebraic structures on the other hand. For example, classes of languages need not be closed under complementation or all preimages under homomorphisms, while monoids can be equipped with a compatible order or they can have a distinguished set of generators. Such generalized varieties and pseudovarieties also have natural counterparts formed by classes of finite (ordered) automata. In this paper the previous approaches are combined. The notion of positive $\mathcal C$-varieties of ordered semiautomata (i.e. no initial and final states are specified) is introduced and their correspondence with positive $\mathcal C$-varieties of languages is proved.

Direct Evidence of Spontaneous Abrikosov Vortex State in Ferromagnetic Superconductor EuFe$_2$(As$_{1-x}$P$_x$)$_2$ with $x=0.21$

Using low-temperature Magnetic Force Microscopy (MFM) we provide direct experimental evidence for spontaneous vortex phase (SVP) formation in EuFe$_2$(As$_{0.79}$P$_{0.21}$)$_2$ single crystal with the superconducting $T^{\rm 0}_{\rm SC}=23.6$~K and ferromagnetic $T_{\rm FM}\sim17.7$~K transition temperatures. Spontaneous vortex-antivortex (V-AV) pairs are imaged in the vicinity of $T_{\rm FM}$. Also, upon cooling cycle near $T_{\rm FM}$ we observe the first-order transition from the short period domain structure, which appears in the Meissner state, into the long period domain structure with spontaneous vortices. It is the first experimental observation of this scenario in the ferromagnetic superconductors. Low-temperature phase is characterized by much larger domains in V-AV state and peculiar branched striped structures at the surface, which are typical for uniaxial ferromagnets with perpendicular magnetic anisotropy (PMA). The domain wall parameters at various temperatures are estimated.

A rank 18 Waring decomposition of $sM_{\langle 3\rangle}$ with 432 symmetries

The recent discovery that the exponent of matrix multiplication is determined by the rank of the symmetrized matrix multiplication tensor has invigorated interest in better understanding symmetrized matrix multiplication. I present an explicit rank 18 Waring decomposition of $sM_{\langle 3\rangle}$ and describe its symmetry group.

The PdBI Arcsecond Whirlpool Survey (PAWS). The Role of Spiral Arms in Cloud and Star Formation

The process that leads to the formation of the bright star forming sites observed along prominent spiral arms remains elusive. We present results of a multi-wavelength study of a spiral arm segment in the nearby grand-design spiral galaxy M51 that belongs to a spiral density wave and exhibits nine gas spurs. The combined observations of the(ionized, atomic, molecular, dusty) interstellar medium (ISM) with star formation tracers (HII regions, young <10Myr stellar clusters) suggest (1) no variation in giant molecular cloud (GMC) properties between arm and gas spurs, (2) gas spurs and extinction feathers arising from the same structure with a close spatial relation between gas spurs and ongoing/recent star formation (despite higher gas surface densities in the spiral arm), (3) no trend in star formation age either along the arm or along a spur, (4) evidence for strong star formation feedback in gas spurs: (5) tentative evidence for star formation triggered by stellar feedback for one spur, and (6) GMC associations (GMAs) being no special entities but the result of blending of gas arm/spur cross-sections in lower resolution observations. We conclude that there is no evidence for a coherent star formation onset mechanism that can be solely associated to the presence of the spiral density wave. This suggests that other (more localized) mechanisms are important to delay star formation such that it occurs in spurs. The evidence of star formation proceeding over several million years within individual spurs implies that the mechanism that leads to star formation acts or is sustained over a longer time-scale.

Higher structure in the unstable Adams spectral sequence

We describe a variant construction of the unstable Adams spectral the sequence for a space $Y$, associated to any free simplicial resolution of $H^*(Y;R)$ for $R=\mathbb{F}_p$ or $\mathbb{Q}$. We use this construction to describe the differentials and filtration in the spectral sequence in terms of appropriate systems of higher cohomology operations.

Comparing Covariate Prioritization via Matching to Machine Learning Methods for Causal Inference using Five Empirical Applications

When investigators seek to estimate causal effects, they often assume that selection into treatment is based only on observed covariates. Under this identification strategy, analysts must adjust for observed confounders. While basic regression models have long been the dominant method of statistical adjustment, more robust methods based on matching or weighting have become more common. Of late, even more flexible methods based on machine learning methods have been developed for statistical adjustment. These machine learning methods are designed to be black box methods with little input from the researcher. Recent research used a data competition to evaluate various methods of statistical adjustment and found that black box methods out performed all other methods of statistical adjustment. Matching methods with covariate prioritization are designed for direct input from substantive investigators in direct contrast to black methods. In this article, we use a different research design to compare matching with covariate prioritization to black box methods. We use black box methods to replicate results from five studies where matching with covariate prioritization was used to customize the statistical adjustment in direct response to substantive expertise. We find little difference across the methods. We conclude with advice for investigators.

Acoustic Impedance Calculation via Numerical Solution of the Inverse Helmholtz Problem

Assigning homogeneous boundary conditions, such as acoustic impedance, to the thermoviscous wave equations (TWE) derived by transforming the linearized Navier-Stokes equations (LNSE) to the frequency domain yields a so-called Helmholtz solver, whose output is a discrete set of complex eigenfunction and eigenvalue pairs. The proposed method -- the inverse Helmholtz solver (iHS) -- reverses such procedure by returning the value of acoustic impedance at one or more unknown impedance boundaries (IBs) of a given domain via spatial integration of the TWE for a given real-valued frequency with assigned conditions on other boundaries. The iHS procedure is applied to a second-order spatial discretization of the TWEs derived on an unstructured grid with staggered grid arrangement. The momentum equation only is extended to the center of each IB face where pressure and velocity components are co-located and treated as unknowns. One closure condition considered for the iHS is the assignment of the surface gradient of pressure phase over the IBs, corresponding to assigning the shape of the acoustic waveform at the IB. The iHS procedure is carried out independently for each frequency in order to return the complete broadband complex impedance distribution at the IBs in any desired frequency range. The iHS approach is first validated against Rott's theory for both inviscid and viscous, rectangular and circular ducts. The impedance of a geometrically complex toy cavity is then reconstructed and verified against companion full compressible unstructured Navier-Stokes simulations resolving the cavity geometry and one-dimensional impedance test tube calculations based on time-domain impedance boundary conditions (TDIBC). The iHS methodology is also shown to capture thermoacoustic effects, with reconstructed impedance values quantitatively in agreement with thermoacoustic growth rates.

Deciphering noise amplification and reduction in open chemical reaction networks

The impact of random fluctuations on the dynamical behavior a complex biological systems is a longstanding issue, whose understanding would shed light on the evolutionary pressure that nature imposes on the intrinsic noise levels and would allow rationally designing synthetic networks with controlled noise. Using the Itō stochastic differential equation formalism, we performed both analytic and numerical analyses of several model systems containing different molecular species in contact with the environment and interacting with each other through mass-action kinetics. These systems represent for example biomolecular oligomerization processes, complex-breakage reactions, signaling cascades or metabolic networks. For chemical reaction networks with zero deficiency values, which admit a detailed- or complex-balanced steady state, all molecular species are uncorrelated. The number of molecules of each species follow a Poisson distribution and their Fano factors, which measure the intrinsic noise, are equal to one. Systems with deficiency one have an unbalanced non-equilibrium steady state and a non-zero S-flux, defined as the flux flowing between the complexes multiplied by an adequate stoichiometric coefficient. In this case, the noise on each species is reduced if the flux flows from the species of lowest to highest complexity, and is amplified is the flux goes in the opposite direction. These results are generalized to systems of deficiency two, which possess two independent non-vanishing S-fluxes, and we conjecture that a similar relation holds for higher deficiency systems.

Many-Body Localization: Stability and Instability

Rare regions with weak disorder (Griffiths regions) have the potential to spoil localization. We describe a non-perturbative construction of local integrals of motion (LIOMs) for a weakly interacting spin chain in one dimension, under a physically reasonable assumption on the statistics of eigenvalues. We discuss ideas about the situation in higher dimensions, where one can no longer ensure that interactions involving the Griffiths regions are much smaller than the typical energy-level spacing for such regions. We argue that ergodicity is restored in dimension d > 1, although equilibration should be extremely slow, similar to the dynamics of glasses.

Fault Detection and Isolation Tools (FDITOOLS) User's Guide

The Fault Detection and Isolation Tools (FDITOOLS) is a collection of MATLAB functions for the analysis and solution of fault detection and model detection problems. The implemented functions are based on the computational procedures described in the Chapters 5, 6 and 7 of the book: "A. Varga, Solving Fault Diagnosis Problems - Linear Synthesis Techniques, Springer, 2017". This document is the User's Guide for the version V1.0 of FDITOOLS. First, we present the mathematical background for solving several basic exact and approximate synthesis problems of fault detection filters and model detection filters. Then, we give in-depth information on the command syntax of the main analysis and synthesis functions. Several examples illustrate the use of the main functions of FDITOOLS.

Complexity of Deciding Detectability in Discrete Event Systems

Detectability of discrete event systems (DESs) is a question whether the current and subsequent states can be determined based on observations. Shu and Lin designed a polynomial-time algorithm to check strong (periodic) detectability and an exponential-time (polynomial-space) algorithm to check weak (periodic) detectability. Zhang showed that checking weak (periodic) detectability is PSpace-complete. This intractable complexity opens a question whether there are structurally simpler DESs for which the problem is tractable. In this paper, we show that it is not the case by considering DESs represented as deterministic finite automata without non-trivial cycles, which are structurally the simplest deadlock-free DESs. We show that even for such very simple DESs, checking weak (periodic) detectability remains intractable. On the contrary, we show that strong (periodic) detectability of DESs can be efficiently verified on a parallel computer.

The Knaster-Tarski theorem versus monotone nonexpansive mappings

Let $X$ be a partially ordered set with the property that each family of order intervals of the form $[a,b],[a,\rightarrow )$ with the finite intersection property has a nonempty intersection. We show that every directed subset of $X$ has a supremum. Then we apply the above result to prove that if $X$ is a topological space with a partial order $\preceq $ for which the order intervals are compact, $\mathcal{F}$ a nonempty commutative family of monotone maps from $X$ into $X$ and there exists $c\in X$ such that $c\preceq Tc$ for every $T\in \mathcal{F}$, then the set of common fixed points of $\mathcal{F}$ is nonempty and has a maximal element. The result, specialized to the case of Banach spaces gives a general fixed point theorem that drops almost all assumptions from the recent results in this area. An application to the theory of integral equations of Urysohn's type is also given.

Efficient methods for computing integrals in electronic structure calculations

Efficient methods are proposed, for computing integrals appeaing in electronic structure calculations. The methods consist of two parts: the first part is to represent the integrals as contour integrals and the second one is to evaluate the contour integrals by the Clenshaw-Curtis quadrature. The efficiency of the proposed methods is demonstrated through numerical experiments.

Diffraction-Aware Sound Localization for a Non-Line-of-Sight Source

We present a novel sound localization algorithm for a non-line-of-sight (NLOS) sound source in indoor environments. Our approach exploits the diffraction properties of sound waves as they bend around a barrier or an obstacle in the scene. We combine a ray tracing based sound propagation algorithm with a Uniform Theory of Diffraction (UTD) model, which simulate bending effects by placing a virtual sound source on a wedge in the environment. We precompute the wedges of a reconstructed mesh of an indoor scene and use them to generate diffraction acoustic rays to localize the 3D position of the source. Our method identifies the convergence region of those generated acoustic rays as the estimated source position based on a particle filter. We have evaluated our algorithm in multiple scenarios consisting of a static and dynamic NLOS sound source. In our tested cases, our approach can localize a source position with an average accuracy error, 0.7m, measured by the L2 distance between estimated and actual source locations in a 7m*7m*3m room. Furthermore, we observe 37% to 130% improvement in accuracy over a state-of-the-art localization method that does not model diffraction effects, especially when a sound source is not visible to the robot.

Jacob's ladders, crossbreeding in the set of $ζ$-factorization formulas and selection of families of $ζ$-kindred real continuous functions

In this paper we introduce the notion of $\zeta$-crossbreeding in a set of $\zeta$-factorization formulas and also the notion of complete hybrid formula as the final result of that crossbreeding. The last formula is used as a criterion for selection of families of $\zeta$-kindred elements in class of real continuous functions. Dedicated to recalling of Gregory Mendel's pea-crossbreeding.

Minimax Estimation of the $L_1$ Distance

We consider the problem of estimating the $L_1$ distance between two discrete probability measures $P$ and $Q$ from empirical data in a nonasymptotic and large alphabet setting. When $Q$ is known and one obtains $n$ samples from $P$, we show that for every $Q$, the minimax rate-optimal estimator with $n$ samples achieves performance comparable to that of the maximum likelihood estimator (MLE) with $n\ln n$ samples. When both $P$ and $Q$ are unknown, we construct minimax rate-optimal estimators whose worst case performance is essentially that of the known $Q$ case with $Q$ being uniform, implying that $Q$ being uniform is essentially the most difficult case. The \emph{effective sample size enlargement} phenomenon, identified in Jiao \emph{et al.} (2015), holds both in the known $Q$ case for every $Q$ and the $Q$ unknown case. However, the construction of optimal estimators for $\|P-Q\|_1$ requires new techniques and insights beyond the approximation-based method of functional estimation in Jiao \emph{et al.} (2015).

Density large deviations for multidimensional stochastic hyperbolic conservation laws

We investigate the density large deviation function for a multidimensional conservation law in the vanishing viscosity limit, when the probability concentrates on weak solutions of a hyperbolic conservation law conservation law. When the conductivity and dif-fusivity matrices are proportional, i.e. an Einstein-like relation is satisfied, the problem has been solved in [4]. When this proportionality does not hold, we compute explicitly the large deviation function for a step-like density profile, and we show that the associated optimal current has a non trivial structure. We also derive a lower bound for the large deviation function, valid for a general weak solution, and leave the general large deviation function upper bound as a conjecture.

mixup: Beyond Empirical Risk Minimization

Large deep neural networks are powerful, but exhibit undesirable behaviors such as memorization and sensitivity to adversarial examples. In this work, we propose mixup, a simple learning principle to alleviate these issues. In essence, mixup trains a neural network on convex combinations of pairs of examples and their labels. By doing so, mixup regularizes the neural network to favor simple linear behavior in-between training examples. Our experiments on the ImageNet-2012, CIFAR-10, CIFAR-100, Google commands and UCI datasets show that mixup improves the generalization of state-of-the-art neural network architectures. We also find that mixup reduces the memorization of corrupt labels, increases the robustness to adversarial examples, and stabilizes the training of generative adversarial networks.

Equality of the usual definitions of Brakke flow

In 1978 Brakke introduced the mean curvature flow in the setting of geometric measure theory. There exist multiple variants of the original definition. Here we prove that most of them are indeed equal. One central point is to correct the proof of Brakke's §3.5, where he develops an estimate for the evolution of the measure of time-dependent test functions.

Dynamic Base Station Repositioning to Improve Spectral Efficiency of Drone Small Cells

With recent advancements in drone technology, researchers are now considering the possibility of deploying small cells served by base stations mounted on flying drones. A major advantage of such drone small cells is that the operators can quickly provide cellular services in areas of urgent demand without having to pre-install any infrastructure. Since the base station is attached to the drone, technically it is feasible for the base station to dynamic reposition itself in response to the changing locations of users for reducing the communication distance, decreasing the probability of signal blocking, and ultimately increasing the spectral efficiency. In this paper, we first propose distributed algorithms for autonomous control of drone movements, and then model and analyse the spectral efficiency performance of a drone small cell to shed new light on the fundamental benefits of dynamic repositioning. We show that, with dynamic repositioning, the spectral efficiency of drone small cells can be increased by nearly 100\% for realistic drone speed, height, and user traffic model and without incurring any major increase in drone energy consumption.

An Unsupervised Homogenization Pipeline for Clustering Similar Patients using Electronic Health Record Data

Electronic health records (EHR) contain a large variety of information on the clinical history of patients such as vital signs, demographics, diagnostic codes and imaging data. The enormous potential for discovery in this rich dataset is hampered by its complexity and heterogeneity. We present the first study to assess unsupervised homogenization pipelines designed for EHR clustering. To identify the optimal pipeline, we tested accuracy on simulated data with varying amounts of redundancy, heterogeneity, and missingness. We identified two optimal pipelines: 1) Multiple Imputation by Chained Equations (MICE) combined with Local Linear Embedding; and 2) MICE, Z-scoring, and Deep Autoencoders.

Deep Neural Network Optimized to Resistive Memory with Nonlinear Current-Voltage Characteristics

Artificial Neural Network computation relies on intensive vector-matrix multiplications. Recently, the emerging nonvolatile memory (NVM) crossbar array showed a feasibility of implementing such operations with high energy efficiency, thus there are many works on efficiently utilizing emerging NVM crossbar array as analog vector-matrix multiplier. However, its nonlinear I-V characteristics restrain critical design parameters, such as the read voltage and weight range, resulting in substantial accuracy loss. In this paper, instead of optimizing hardware parameters to a given neural network, we propose a methodology of reconstructing a neural network itself optimized to resistive memory crossbar arrays. To verify the validity of the proposed method, we simulated various neural network with MNIST and CIFAR-10 dataset using two different specific Resistive Random Access Memory (RRAM) model. Simulation results show that our proposed neural network produces significantly higher inference accuracies than conventional neural network when the synapse devices have nonlinear I-V characteristics.

Rate-Distortion Region of a Gray-Wyner Model with Side Information

In this work, we establish a full single-letter characterization of the rate-distortion region of an instance of the Gray-Wyner model with side information at the decoders. Specifically, in this model an encoder observes a pair of memoryless, arbitrarily correlated, sources $(S^n_1,S^n_2)$ and communicates with two receivers over an error-free rate-limited link of capacity $R_0$, as well as error-free rate-limited individual links of capacities $R_1$ to the first receiver and $R_2$ to the second receiver. Both receivers reproduce the source component $S^n_2$ losslessly; and Receiver $1$ also reproduces the source component $S^n_1$ lossily, to within some prescribed fidelity level $D_1$. Also, Receiver $1$ and Receiver $2$ are equipped respectively with memoryless side information sequences $Y^n_1$ and $Y^n_2$. Important in this setup, the side information sequences are arbitrarily correlated among them, and with the source pair $(S^n_1,S^n_2)$; and are not assumed to exhibit any particular ordering. Furthermore, by specializing the main result to two Heegard-Berger models with successive refinement and scalable coding, we shed light on the roles of the common and private descriptions that the encoder should produce and what they should carry optimally. We develop intuitions by analyzing the developed single-letter optimal rate-distortion regions of these models, and discuss some insightful binary examples.

Fourier-based numerical approximation of the Weertman equation for moving dislocations

This work discusses the numerical approximation of a nonlinear reaction-advection-diffusion equation, which is a dimensionless form of the Weertman equation. This equation models steadily-moving dislocations in materials science. It reduces to the celebrated Peierls-Nabarro equation when its advection term is set to zero. The approach rests on considering a time-dependent formulation, which admits the equation under study as its long-time limit. Introducing a Preconditioned Collocation Scheme based on Fourier transforms, the iterative numerical method presented solves the time-dependent problem, delivering at convergence the desired numerical solution to the Weertman equation. Although it rests on an explicit time-evolution scheme, the method allows for large time steps, and captures the solution in a robust manner. Numerical results illustrate the efficiency of the approach for several types of nonlinearities.

Design Decisions for Weave: A Real-Time Web-based Collaborative Visualization Framework

There are many web-based visualization systems available to date, each having its strengths and limitations. The goals these systems set out to accomplish influence design decisions and determine how reusable and scalable they are. Weave is a new web-based visualization platform with the broad goal of enabling visualization of any available data by anyone for any purpose. Our open source framework supports highly interactive linked visualizations for users of varying skill levels. What sets Weave apart from other systems is its consideration for real-time remote collaboration with session history. We provide a detailed account of the various framework designs we considered with comparisons to existing state-of-the-art systems.

Suzaku Analysis of the Supernova Remnant G306.3-0.9 and the Gamma-ray View of Its Neighborhood

We present an investigation of the supernova remnant (SNR) G306.3$-$0.9 using archival multi-wavelength data. The Suzaku spectra are well described by two-component thermal plasma models: The soft component is in ionization equilibrium and has a temperature $\sim$0.59 keV, while the hard component has temperature $\sim$3.2 keV and ionization time-scale $\sim$$2.6\times10^{10}$ cm$^{-3}$ s. We clearly detected Fe K-shell line at energy of $\sim$6.5 keV from this remnant. The overabundances of Si, S, Ar, Ca, and Fe confirm that the X-ray emission has an ejecta origin. The centroid energy of the Fe-K line supports that G306.3$-$0.9 is a remnant of a Type Ia supernova (SN) rather than a core-collapse SN. The GeV gamma-ray emission from G306.3$-$0.9 and its surrounding were analyzed using about 6 years of Fermi data. We report about the non-detection of G306.3$-$0.9 and the detection of a new extended gamma-ray source in the south-west of G306.3$-$0.9 with a significance of $\sim$13$\sigma$. We discuss several scenarios for these results with the help of data from other wavebands to understand the SNR and its neighborhood.

Japanese Sentiment Classification using a Tree-Structured Long Short-Term Memory with Attention

Previous approaches to training syntax-based sentiment classification models required phrase-level annotated corpora, which are not readily available in many languages other than English. Thus, we propose the use of tree-structured Long Short-Term Memory with an attention mechanism that pays attention to each subtree of the parse tree. Experimental results indicate that our model achieves the state-of-the-art performance in a Japanese sentiment classification task.

Covariances, Robustness, and Variational Bayes

Mean-field Variational Bayes (MFVB) is an approximate Bayesian posterior inference technique that is increasingly popular due to its fast runtimes on large-scale datasets. However, even when MFVB provides accurate posterior means for certain parameters, it often mis-estimates variances and covariances. Furthermore, prior robustness measures have remained undeveloped for MFVB. By deriving a simple formula for the effect of infinitesimal model perturbations on MFVB posterior means, we provide both improved covariance estimates and local robustness measures for MFVB, thus greatly expanding the practical usefulness of MFVB posterior approximations. The estimates for MFVB posterior covariances rely on a result from the classical Bayesian robustness literature relating derivatives of posterior expectations to posterior covariances and include the Laplace approximation as a special case. Our key condition is that the MFVB approximation provides good estimates of a select subset of posterior means---an assumption that has been shown to hold in many practical settings. In our experiments, we demonstrate that our methods are simple, general, and fast, providing accurate posterior uncertainty estimates and robustness measures with runtimes that can be an order of magnitude faster than MCMC.

Are multi-factor Gaussian term structure models still useful? An empirical analysis on Italian BTPs

In this paper, we empirically study models for pricing Italian sovereign bonds under a reduced form framework, by assuming different dynamics for the short-rate process. We analyze classical Cox-Ingersoll-Ross and Vasicek multi-factor models, with a focus on optimization algorithms applied in the calibration exercise. The Kalman filter algorithm together with a maximum likelihood estimation method are considered to fit the Italian term-structure over a 12-year horizon, including the global financial crisis and the euro area sovereign debt crisis. Analytic formulas for the gradient vector and the Hessian matrix of the likelihood function are provided.

Probing valley filtering effect by Andreev reflection in zigzag graphene nanoribbon

Ballistic point contact (BPC) with zigzag edges in graphene is a main candidate of a valley filter, in which the polarization of the valley degree of freedom can be selected by using a local gate voltage. Here, we propose to detect the valley filtering effect by Andreev reflection. Because electrons in the lowest conduction band and the highest valence band of the BPC possess opposite chirality, the inter-band Andreev reflection is strongly suppressed, after multiple scattering and interference. We draw this conclusion by both the scattering matrix analysis and the numerical simulation. The Andreev reflection as a function of the incident energy of electrons and the local gate voltage at the BPC is obtained, by which the parameter region for a perfect valley filter and the direction of valley polarization can be determined. The Andreev reflection exhibits an oscillatory decay with the length of the BPC, indicating a negative correlation to valley polarization.

Generalized Approximate Message-Passing Decoder for Universal Sparse Superposition Codes

Sparse superposition (SS) codes were originally proposed as a capacity-achieving communication scheme over the additive white Gaussian noise channel (AWGNC) [1]. Very recently, it was discovered that these codes are universal, in the sense that they achieve capacity over any memoryless channel under generalized approximate message-passing (GAMP) decoding [2], although this decoder has never been stated for SS codes. In this contribution we introduce the GAMP decoder for SS codes, we confirm empirically the universality of this communication scheme through its study on various channels and we provide the main analysis tools: state evolution and potential. We also compare the performance of GAMP with the Bayes-optimal MMSE decoder. We empirically illustrate that despite the presence of a phase transition preventing GAMP to reach the optimal performance, spatial coupling allows to boost the performance that eventually tends to capacity in a proper limit. We also prove that, in contrast with the AWGNC case, SS codes for binary input channels have a vanishing error floor in the limit of large codewords. Moreover, the performance of Hadamard-based encoders is assessed for practical implementations.

LAAIR: A Layered Architecture for Autonomous Interactive Robots

When developing general purpose robots, the overarching software architecture can greatly affect the ease of accomplishing various tasks. Initial efforts to create unified robot systems in the 1990s led to hybrid architectures, emphasizing a hierarchy in which deliberative plans direct the use of reactive skills. However, since that time there has been significant progress in the low-level skills available to robots, including manipulation and perception, making it newly feasible to accomplish many more tasks in real-world domains. There is thus renewed optimism that robots will be able to perform a wide array of tasks while maintaining responsiveness to human operators. However, the top layer in traditional hybrid architectures, designed to achieve long-term goals, can make it difficult to react quickly to human interactions during goal-driven execution. To mitigate this difficulty, we propose a novel architecture that supports such transitions by adding a top-level reactive module which has flexible access to both reactive skills and a deliberative control module. To validate this architecture, we present a case study of its application on a domestic service robot platform.

3D Human Pose Estimation in RGBD Images for Robotic Task Learning

We propose an approach to estimate 3D human pose in real world units from a single RGBD image and show that it exceeds performance of monocular 3D pose estimation approaches from color as well as pose estimation exclusively from depth. Our approach builds on robust human keypoint detectors for color images and incorporates depth for lifting into 3D. We combine the system with our learning from demonstration framework to instruct a service robot without the need of markers. Experiments in real world settings demonstrate that our approach enables a PR2 robot to imitate manipulation actions observed from a human teacher.

Simultaneous non-vanishing for Dirichlet L-functions

We extend the work of Fouvry, Kowalski and Michel on correlation between Hecke eigenvalues of modular forms and algebraic trace functions in order to establish an asymptotic formula for a generalized cubic moment of modular L-functions at the central point s = 1/2 and for prime moduli q. As an application, we exploit our recent result on the mollification of the fourth moment of Dirichlet L-functions to derive that for any pair $(\omega_1,\omega_2)$ of multiplicative characters modulo q, there is a positive proportion of $\chi$ (mod q) such that $L(\chi, 1/2 ), L(\chi\omega_1, 1/2 )$ and $L(\chi\omega_2, 1/2)$ are simultaneously not too small.

Wehrl Entropy Based Quantification of Nonclassicality for Single Mode Quantum Optical States

Nonclassical states of a quantized light are described in terms of Glauber-Sudarshan P distribution which is not a genuine classical probability distribution. Despite several attempts, defining a uniform measure of nonclassicality (NC) for the single mode quantum states of light is yet an open task. In our previous work [Phys. Rev. A 95, 012330 (2017)] we have shown that the existing well-known measures fail to quantify the NC of single mode states that are generated under multiple NC-inducing operations. Recently, Ivan et. al. [Quantum. Inf. Process. 11, 853 (2012)] have defined a measure of non-Gaussian character of quantum optical states in terms of Wehrl entropy. Here, we adopt this concept in the context of single mode NC. In this paper, we propose a new quantification of NC for the single mode quantum states of light as the difference between the total Wehrl entropy of the state and the maximum Wehrl entropy arising due to its classical characteristics. This we achieve by subtracting from its Wehrl entropy, the maximum Wehrl entropy attainable by any classical state that has same randomness as measured in terms of von-Neumann entropy. We obtain analytic expressions of NC for most of the states, in particular, all pure states and Gaussian mixed states. However, the evaluation of NC for the non-Gaussian mixed states is subject to extensive numerical computation that lies beyond the scope of the current work. We show that, along with the states generated under single NC-inducing operations, also for the broader class of states that are generated under multiple NC-inducing operations, our quantification enumerates the NC consistently.

Attention-based Natural Language Person Retrieval

Following the recent progress in image classification and captioning using deep learning, we develop a novel natural language person retrieval system based on an attention mechanism. More specifically, given the description of a person, the goal is to localize the person in an image. To this end, we first construct a benchmark dataset for natural language person retrieval. To do so, we generate bounding boxes for persons in a public image dataset from the segmentation masks, which are then annotated with descriptions and attributes using the Amazon Mechanical Turk. We then adopt a region proposal network in Faster R-CNN as a candidate region generator. The cropped images based on the region proposals as well as the whole images with attention weights are fed into Convolutional Neural Networks for visual feature extraction, while the natural language expression and attributes are input to Bidirectional Long Short- Term Memory (BLSTM) models for text feature extraction. The visual and text features are integrated to score region proposals, and the one with the highest score is retrieved as the output of our system. The experimental results show significant improvement over the state-of-the-art method for generic object retrieval and this line of research promises to benefit search in surveillance video footage.

Large Scale Automated Forecasting for Monitoring Network Safety and Security

Real time large scale streaming data pose major challenges to forecasting, in particular defying the presence of human experts to perform the corresponding analysis. We present here a class of models and methods used to develop an automated, scalable and versatile system for large scale forecasting oriented towards safety and security monitoring. Our system provides short and long term forecasts and uses them to detect safety and security issues in relation with multiple internet connected devices well in advance they might take place.

Contextual Regression: An Accurate and Conveniently Interpretable Nonlinear Model for Mining Discovery from Scientific Data

Machine learning algorithms such as linear regression, SVM and neural network have played an increasingly important role in the process of scientific discovery. However, none of them is both interpretable and accurate on nonlinear datasets. Here we present contextual regression, a method that joins these two desirable properties together using a hybrid architecture of neural network embedding and dot product layer. We demonstrate its high prediction accuracy and sensitivity through the task of predictive feature selection on a simulated dataset and the application of predicting open chromatin sites in the human genome. On the simulated data, our method achieved high fidelity recovery of feature contributions under random noise levels up to 200%. On the open chromatin dataset, the application of our method not only outperformed the state of the art method in terms of accuracy, but also unveiled two previously unfound open chromatin related histone marks. Our method can fill the blank of accurate and interpretable nonlinear modeling in scientific data mining tasks.

Multi-time correlators in continuous measurement of qubit observables

We consider multi-time correlators for output signals from linear detectors, continuously measuring several qubit observables at the same time. Using the quantum Bayesian formalism, we show that for unital (symmetric) evolution in the absence of phase backaction, an $N$-time correlator can be expressed as a product of two-time correlators when $N$ is even. For odd $N$, there is a similar factorization, which also includes a single-time average. Theoretical predictions agree well with experimental results for two detectors, which simultaneously measure non-commuting qubit observables.

Parallelism, Concurrency and Distribution in Constraint Handling Rules: A Survey

Constraint Handling Rules is an effective concurrent declarative programming language and a versatile computational logic formalism. CHR programs consist of guarded reactive rules that transform multisets of constraints. One of the main features of CHR is its inherent concurrency. Intuitively, rules can be applied to parts of a multiset in parallel. In this comprehensive survey, we give an overview of concurrent and parallel as well as distributed CHR semantics, standard and more exotic, that have been proposed over the years at various levels of refinement. These semantics range from the abstract to the concrete. They are related by formal soundness results. Their correctness is established as correspondence between parallel and sequential computations. We present common concise sample CHR programs that have been widely used in experiments and benchmarks. We review parallel CHR implementations in software and hardware. The experimental results obtained show a consistent parallel speedup. Most implementations are available online. The CHR formalism can also be used to implement and reason with models for concurrency. To this end, the Software Transaction Model, the Actor Model, Colored Petri Nets and the Join-Calculus have been faithfully encoded in CHR. Under consideration in Theory and Practice of Logic Programming (TPLP).

Robustness against the channel effect in pathological voice detection

Many people are suffering from voice disorders, which can adversely affect the quality of their lives. In response, some researchers have proposed algorithms for automatic assessment of these disorders, based on voice signals. However, these signals can be sensitive to the recording devices. Indeed, the channel effect is a pervasive problem in machine learning for healthcare. In this study, we propose a detection system for pathological voice, which is robust against the channel effect. This system is based on a bidirectional LSTM network. To increase the performance robustness against channel mismatch, we integrate domain adversarial training (DAT) to eliminate the differences between the devices. When we train on data recorded on a high-quality microphone and evaluate on smartphone data without labels, our robust detection system increases the PR-AUC from 0.8448 to 0.9455 (and 0.9522 with target sample labels). To the best of our knowledge, this is the first study applying unsupervised domain adaptation to pathological voice detection. Notably, our system does not need target device sample labels, which allows for generalization to many new devices.

An Effective Framework for Constructing Exponent Lattice Basis of Nonzero Algebraic Numbers

Computing a basis for the exponent lattice of algebraic numbers is a basic problem in the field of computational number theory with applications to many other areas. The main cost of a well-known algorithm \cite{ge1993algorithms,kauers2005algorithms} solving the problem is on computing the primitive element of the extended field generated by the given algebraic numbers. When the extended field is of large degree, the problem seems intractable by the tool implementing the algorithm. In this paper, a special kind of exponent lattice basis is introduced. An important feature of the basis is that it can be inductively constructed, which allows us to deal with the given algebraic numbers one by one when computing the basis. Based on this, an effective framework for constructing exponent lattice basis is proposed. Through computing a so-called pre-basis first and then solving some linear Diophantine equations, the basis can be efficiently constructed. A new certificate for multiplicative independence and some techniques for decreasing degrees of algebraic numbers are provided to speed up the computation. The new algorithm has been implemented with Mathematica and its effectiveness is verified by testing various examples. Moreover, the algorithm is applied to program verification for finding invariants of linear loops.

Competing evolutionary paths in growing populations with applications to multidrug resistance

Investigating the emergence of a particular cell type is a recurring theme in models of growing cellular populations. The evolution of resistance to therapy is a classic example. Common questions are: when does the cell type first occur, and via which sequence of steps is it most likely to emerge? For growing populations, these questions can be formulated in a general framework of branching processes spreading through a graph from a root to a target vertex. Cells have a particular fitness value on each vertex and can transition along edges at specific rates. Vertices represents cell states, say \mic{genotypes }or physical locations, while possible transitions are acquiring a mutation or cell migration. We focus on the setting where cells at the root vertex have the highest fitness and transition rates are small. Simple formulas are derived for the time to reach the target vertex and for the probability that it is reached along a given path in the graph. We demonstrate our results on \mic{several scenarios relevant to the emergence of drug resistance}, including: the orderings of resistance-conferring mutations in bacteria and the impact of imperfect drug penetration in cancer.

Transient flows in active porous media

Stimuli-responsive materials that modify their shape in response to changes in environmental conditions -- such as solute concentration, temperature, pH, and stress -- are widespread in nature and technology. Applications include micro- and nanoporous materials used in filtration and flow control. The physiochemical mechanisms that induce internal volume modifications have been widely studies. The coupling between induced volume changes and solute transport through porous materials, however, is not well understood. Here, we consider advective and diffusive transport through a small channel linking two large reservoirs. A section of stimulus-responsive material regulates the channel permeability, which is a function of the local solute concentration. We derive an exact solution to the coupled transport problem and demonstrate the existence of a flow regime in which the steady state is reached via a damped oscillation around the equilibrium concentration value. Finally, the feasibility of an experimental observation of the phenomena is discussed. Please note that this version of the paper has not been formally peer reviewed, revised or accepted by a journal.

An information model for modular robots: the Hardware Robot Information Model (HRIM)

Today's landscape of robotics is dominated by vertical integration where single vendors develop the final product leading to slow progress, expensive products and customer lock-in. Opposite to this, an horizontal integration would result in a rapid development of cost-effective mass-market products with an additional consumer empowerment. The transition of an industry from vertical integration to horizontal integration is typically catalysed by de facto industry standards that enable a simplified and seamless integration of products. However, in robotics there is currently no leading candidate for a global plug-and-play standard. This paper tackles the problem of incompatibility between robot components that hinder the reconfigurability and flexibility demanded by the robotics industry. Particularly, it presents a model to create plug-and-play robot hardware components. Rather than iteratively evolving previous ontologies, our proposed model answers the needs identified by the industry while facilitating interoperability, measurability and comparability of robotics technology. Our approach differs significantly with the ones presented before as it is hardware-oriented and establishes a clear set of actions towards the integration of this model in real environments and with real manufacturers.

Detecting Adversarial Samples Using Density Ratio Estimates

Machine learning models, especially based on deep architectures are used in everyday applications ranging from self driving cars to medical diagnostics. It has been shown that such models are dangerously susceptible to adversarial samples, indistinguishable from real samples to human eye, adversarial samples lead to incorrect classifications with high confidence. Impact of adversarial samples is far-reaching and their efficient detection remains an open problem. We propose to use direct density ratio estimation as an efficient model agnostic measure to detect adversarial samples. Our proposed method works equally well with single and multi-channel samples, and with different adversarial sample generation methods. We also propose a method to use density ratio estimates for generating adversarial samples with an added constraint of preserving density ratio.

The Query Complexity of Cake Cutting

We study the query complexity of cake cutting and give lower and upper bounds for computing approximately envy-free, perfect, and equitable allocations with the minimum number of cuts. The lower bounds are tight for computing connected envy-free allocations among n=3 players and for computing perfect and equitable allocations with minimum number of cuts between n=2 players. We also formalize moving knife procedures and show that a large subclass of this family, which captures all the known moving knife procedures, can be simulated efficiently with arbitrarily small error in the Robertson-Webb query model.

Stacked Convolutional and Recurrent Neural Networks for Music Emotion Recognition

This paper studies the emotion recognition from musical tracks in the 2-dimensional valence-arousal (V-A) emotional space. We propose a method based on convolutional (CNN) and recurrent neural networks (RNN), having significantly fewer parameters compared with the state-of-the-art method for the same task. We utilize one CNN layer followed by two branches of RNNs trained separately for arousal and valence. The method was evaluated using the 'MediaEval2015 emotion in music' dataset. We achieved an RMSE of 0.202 for arousal and 0.268 for valence, which is the best result reported on this dataset.

Timed Automata with Polynomial Delay and their Expressiveness

We consider previous models of Timed, Probabilistic and Stochastic Timed Automata, we introduce our model of Timed Automata with Polynomial Delay and we characterize the expressiveness of these models relative to each other.

Superconducting properties of Cu intercalated Bi$_2$Se$_3$ studied by Muon Spin Spectroscopy

We present muon spin rotation measurements on superconducting Cu intercalated Bi$_2$Se$_3$, which was suggested as a realization of a topological superconductor. We observe a clear evidence of the superconducting transition below 4 K, where the width of magnetic field distribution increases as the temperature is decreased. The measured broadening at mK temperatures suggests a large London penetration depth in the $ab$ plane ($\lambda_{\mathrm{eff}}\sim 1.6$ $\mathrm{\mu}$m). We show that the temperature dependence of this broadening follows the BCS prediction, but could be consistent with several gap symmetries.

Time-domain THz spectroscopy reveals coupled protein-hydration dielectric response in solutions of native and fibrils of human lyso-zyme

Here we reveal details of the interaction between human lysozyme proteins, both native and fibrils, and their water environment by intense terahertz time domain spectroscopy. With the aid of a rigorous dielectric model, we determine the amplitude and phase of the oscillating dipole induced by the THz field in the volume containing the protein and its hydration water. At low concentrations, the amplitude of this induced dipolar response decreases with increasing concentration. Beyond a certain threshold, marking the onset of the interactions between the extended hydration shells, the amplitude remains fixed but the phase of the induced dipolar response, which is initially in phase with the applied THz field, begins to change. The changes observed in the THz response reveal protein-protein interactions me-diated by extended hydration layers, which may control fibril formation and may have an important role in chemical recognition phenomena.

Inversion of Qubit Energy Levels in Qubit-Oscillator Circuits in the Deep-Strong-Coupling Regime

We report on experimentally measured light shifts of superconducting flux qubits deep-strongly coupled to LC oscillators, where the coupling constants are comparable to the qubit and oscillator resonance frequencies. By using two-tone spectroscopy, the energies of the six lowest levels of each circuit are determined. We find huge Lamb shifts that exceed 90% of the bare qubit frequencies and inversions of the qubits' ground and excited states when there are a finite number of photons in the oscillator. Our experimental results agree with theoretical predictions based on the quantum Rabi model.

Deep Multiple Instance Feature Learning via Variational Autoencoder

We describe a novel weakly supervised deep learning framework that combines both the discriminative and generative models to learn meaningful representation in the multiple instance learning (MIL) setting. MIL is a weakly supervised learning problem where labels are associated with groups of instances (referred as bags) instead of individual instances. To address the essential challenge in MIL problems raised from the uncertainty of positive instances label, we use a discriminative model regularized by variational autoencoders (VAEs) to maximize the differences between latent representations of all instances and negative instances. As a result, the hidden layer of the variational autoencoder learns meaningful representation. This representation can effectively be used for MIL problems as illustrated by better performance on the standard benchmark datasets comparing to the state-of-the-art approaches. More importantly, unlike most related studies, the proposed framework can be easily scaled to large dataset problems, as illustrated by the audio event detection and segmentation task. Visualization also confirms the effectiveness of the latent representation in discriminating positive and negative classes.

Regularity of envelopes in Kähler classes

We establish the C^{1,1} regularity of quasi-psh envelopes in a Kahler class, confirming a conjecture of Berman.

$S^1$-equivariant Index theorems and Morse inequalities on complex manifolds with boundary

Let $M$ be a complex manifold of dimension $n$ with smooth connected boundary $X$. Assume that $\overline M$ admits a holomorphic $S^1$-action preserving the boundary $X$ and the $S^1$-action is transversal and CR on $X$. We show that the $\overline\partial$-Neumann Laplacian on $M$ is transversally elliptic and as a consequence, the $m$-th Fourier component of the $q$-th Dolbeault cohomology group $H^q_m(\overline M)$ is finite dimensional, for every $m\in\mathbb Z$ and every $q=0,1,\ldots,n$. This enables us to define $\sum^{n}_{j=0}(-1)^j{\rm dim\,}H^q_m(\overline M)$ the $m$-th Fourier component of the Euler characteristic on $M$ and to study large $m$-behavior of $H^q_m(\overline M)$. In this paper, we establish an index formula for $\sum^{n}_{j=0}(-1)^j{\rm dim\,}H^q_m(\overline M)$ and Morse inequalities for $H^q_m(\overline M)$.

Internal Model from Observations for Reward Shaping

Reinforcement learning methods require careful design involving a reward function to obtain the desired action policy for a given task. In the absence of hand-crafted reward functions, prior work on the topic has proposed several methods for reward estimation by using expert state trajectories and action pairs. However, there are cases where complete or good action information cannot be obtained from expert demonstrations. We propose a novel reinforcement learning method in which the agent learns an internal model of observation on the basis of expert-demonstrated state trajectories to estimate rewards without completely learning the dynamics of the external environment from state-action pairs. The internal model is obtained in the form of a predictive model for the given expert state distribution. During reinforcement learning, the agent predicts the reward as a function of the difference between the actual state and the state predicted by the internal model. We conducted multiple experiments in environments of varying complexity, including the Super Mario Bros and Flappy Bird games. We show our method successfully trains good policies directly from expert game-play videos.

Characterizations of quasitrivial symmetric nondecreasing associative operations

In this paper we are interested in the class of n-ary operations on an arbitrary chain that are quasitrivial, symmetric, nondecreasing, and associative. We first provide a description of these operations. We then prove that associativity can be replaced with bisymmetry in the definition of this class. Finally we investigate the special situation where the chain is finite.

Multivariate Dependency Measure based on Copula and Gaussian Kernel

We propose a new multivariate dependency measure. It is obtained by considering a Gaussian kernel based distance between the copula transform of the given d-dimensional distribution and the uniform copula and then appropriately normalizing it. The resulting measure is shown to satisfy a number of desirable properties. A nonparametric estimate is proposed for this dependency measure and its properties (finite sample as well as asymptotic) are derived. Some comparative studies of the proposed dependency measure estimate with some widely used dependency measure estimates on artificial datasets are included. A non-parametric test of independence between two or more random variables based on this measure is proposed. A comparison of the proposed test with some existing nonparametric multivariate test for independence is presented.

The nature of the tensor order in Cd2Re2O7

The pyrochlore metal Cd2Re2O7 has been recently investigated by second-harmonic generation (SHG) reflectivity. In this paper, we develop a general formalism that allows for the identification of the relevant tensor components of the SHG from azimuthal scans. We demonstrate that the secondary order parameter identified by SHG at the structural phase transition is the x2-y2 component of the axial toroidal quadrupole. This differs from the 3z2-r2 symmetry of the atomic displacements associated with the I-4m2 crystal structure that was previously thought to be its origin. Within the same formalism, we suggest that the primary order parameter detected in the SHG experiment is the 3z2-r2 component of the magnetic quadrupole. We discuss the general mechanism driving the phase transition in our proposed framework, and suggest experiments, particularly resonant X-ray scattering ones, that could clarify this issue.

Efficient and consistent inference of ancestral sequences in an evolutionary model with insertions and deletions under dense taxon sampling

In evolutionary biology, the speciation history of living organisms is represented graphically by a phylogeny, that is, a rooted tree whose leaves correspond to current species and branchings indicate past speciation events. Phylogenies are commonly estimated from molecular sequences, such as DNA sequences, collected from the species of interest. At a high level, the idea behind this inference is simple: the further apart in the Tree of Life are two species, the greater is the number of mutations to have accumulated in their genomes since their most recent common ancestor. In order to obtain accurate estimates in phylogenetic analyses, it is standard practice to employ statistical approaches based on stochastic models of sequence evolution on a tree. For tractability, such models necessarily make simplifying assumptions about the evolutionary mechanisms involved. In particular, commonly omitted are insertions and deletions of nucleotides -- also known as indels. Properly accounting for indels in statistical phylogenetic analyses remains a major challenge in computational evolutionary biology. Here we consider the problem of reconstructing ancestral sequences on a known phylogeny in a model of sequence evolution incorporating nucleotide substitutions, insertions and deletions, specifically the classical TKF91 process. We focus on the case of dense phylogenies of bounded height, which we refer to as the taxon-rich setting, where statistical consistency is achievable. We give the first polynomial-time ancestral reconstruction algorithm with provable guarantees under constant rates of mutation. Our algorithm succeeds when the phylogeny satisfies the "big bang" condition, a necessary and sufficient condition for statistical consistency in this context.

Flow Characteristics and Cores of Complex Network and Multiplex Type Systems

Subject of research is complex networks and network systems. The network system is defined as a complex network in which flows are moved. Classification of flows in the network is carried out on the basis of ordering and continuity. It is shown that complex networks with different types of flows generate various network systems. Flow analogues of the basic concepts of the theory of complex networks are introduced and the main problems of this theory in terms of flow characteristics are formulated. Local and global flow characteristics of networks bring closer the theory of complex networks to the systems theory and systems analysis. Concept of flow core of network system is introduced and defined how it simplifies the process of its investigation. Concepts of kernel and flow core of multiplex are determined. Features of operation of multiplex type systems are analyzed.

Pattern-forming fronts in a Swift-Hohenberg equation with directional quenching - parallel and oblique stripes

We study the effect of domain growth on the orientation of striped phases in a Swift-Hohenberg equation. Domain growth is encoded in a step-like parameter dependence that allows stripe formation in a half plane, and suppresses patterns in the complement, while the boundary of the pattern-forming region is propagating with fixed normal velocity. We construct front solutions that leave behind stripes in the pattern-forming region that are parallel to or at a small oblique angle to the boundary. Technically, the construction of stripe formation parallel to the boundary relies on ill-posed, infinite-dimensional spatial dynamics. Stripes forming at a small oblique angle are constructed using a functional-analytic, perturbative approach. Here, the main difficulties are the presence of continuous spectrum and the fact that small oblique angles appear as a singular perturbation in a traveling-wave problem. We resolve the former difficulty using a farfield-core decomposition and Fredholm theory in weighted spaces. The singular perturbation problem is resolved using preconditioners and boot-strapping.

Generalized Minimum Distance Estimators in Linear Regression with Dependent Errors

This paper discusses minimum distance estimation method in the linear regression model with dependent errors which are strongly mixing. The regression parameters are estimated through the minimum distance estimation method, and asymptotic distributional properties of the estimators are discussed. A simulation study compares the performance of the minimum distance estimator with other well celebrated estimator. This simulation study shows the superiority of the minimum distance estimator over another estimator. KoulMde (R package) which was used for the simulation study is available online. See section 4 for the detail.

Live Service Migration in Mobile Edge Clouds

Mobile edge clouds (MECs) bring the benefits of the cloud closer to the user, by installing small cloud infrastructures at the network edge. This enables a new breed of real-time applications, such as instantaneous object recognition and safety assistance in intelligent transportation systems, that require very low latency. One key issue that comes with proximity is how to ensure that users always receive good performance as they move across different locations. Migrating services between MECs is seen as the means to achieve this. This article presents a layered framework for migrating active service applications that are encapsulated either in virtual machines (VMs) or containers. This layering approach allows a substantial reduction in service downtime. The framework is easy to implement using readily available technologies, and one of its key advantages is that it supports containers, which is a promising emerging technology that offers tangible benefits over VMs. The migration performance of various real applications is evaluated by experiments under the presented framework. Insights drawn from the experimentation results are discussed.

Induced density correlations in a sonic black hole condensate

Analog black/white hole pairs, consisting of a region of supersonic flow, have been achieved in a recent experiment by J. Steinhauer using an elongated Bose-Einstein condensate. A growing standing density wave, and a checkerboard feature in the density-density correlation function, were observed in the supersonic region. We model the density-density correlation function, taking into account both quantum fluctuations and the shot-to-shot variation of atom number normally present in ultracold-atom experiments. We find that quantum fluctuations alone produce some, but not all, of the features of the correlation function, whereas atom-number fluctuation alone can produce all the observed features, and agreement is best when both are included. In both cases, the density-density correlation is not intrinsic to the fluctuations, but rather is induced by modulation of the standing wave caused by the fluctuations.

Genus growth in $\mathbb{Z}_p$-towers of function fields

Let $K$ be a function field over a finite field $k$ of characteristic $p$ and let $K_{\infty}/K$ be a geometric extension with Galois group $\mathbb{Z}_p$. Let $K_n$ be the corresponding subextension with Galois group $\mathbb{Z}/p^n\mathbb{Z}$ and genus $g_n$. In this paper, we give a simple explicit formula $g_n$ in terms of an explicit Witt vector construction of the $\mathbb{Z}_p$-tower. This formula leads to a tight lower bound on $g_n$ which is quadratic in $p^n$. Furthermore, we determine all $\mathbb{Z}_p$-towers for which the genus sequence is stable, in the sense that there are $a,b,c \in \mathbb{Q}$ such that $g_n=a p^{2n}+b p^n +c$ for $n$ large enough. Such genus stable towers are expected to have strong stable arithmetic properties for their zeta functions. A key technical contribution of this work is a new simplified formula for the Schmid-Witt symbol coming from local class field theory.

Topological Phases emerging from Spin-Orbital Physics

We study the evolution of spin-orbital correlations in an inhomogeneous quantum system with an impurity replacing a doublon by a holon orbital degree of freedom. Spin-orbital entanglement is large when spin correlations are antiferromagnetic, while for a ferromagnetic host we obtain a pure orbital description. In this regime the orbital model can be mapped on spinless fermions and we uncover topological phases with zero energy modes at the edge or at the domain between magnetically inequivalent regions.

Accurate and Diverse Sampling of Sequences based on a "Best of Many" Sample Objective

For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence. This problem has been formalized as a sequence extrapolation problem, where a number of observations are used to predict the sequence into the future. Real-world scenarios demand a model of uncertainty of such predictions, as predictions become increasingly uncertain -- in particular on long time horizons. While impressive results have been shown on point estimates, scenarios that induce multi-modal distributions over future sequences remain challenging. Our work addresses these challenges in a Gaussian Latent Variable model for sequence prediction. Our core contribution is a "Best of Many" sample objective that leads to more accurate and more diverse predictions that better capture the true variations in real-world sequence data. Beyond our analysis of improved model fit, our models also empirically outperform prior work on three diverse tasks ranging from traffic scenes to weather data.

Exploring RNN-Transducer for Chinese Speech Recognition

End-to-end approaches have drawn much attention recently for significantly simplifying the construction of an automatic speech recognition (ASR) system. RNN transducer (RNN-T) is one of the popular end-to-end methods. Previous studies have shown that RNN-T is difficult to train and a very complex training process is needed for a reasonable performance. In this paper, we explore RNN-T for a Chinese large vocabulary continuous speech recognition (LVCSR) task and aim to simplify the training process while maintaining performance. First, a new strategy of learning rate decay is proposed to accelerate the model convergence. Second, we find that adding convolutional layers at the beginning of the network and using ordered data can discard the pre-training process of the encoder without loss of performance. Besides, we design experiments to find a balance among the usage of GPU memory, training circle and model performance. Finally, we achieve 16.9% character error rate (CER) on our test set which is 2% absolute improvement from a strong BLSTM CE system with language model trained on the same text corpus.

A Debt-Aware Learning Approach for Resource Adaptations in Cloud Elasticity Management

Elasticity is a cloud property that enables applications and its execution systems to dynamically acquire and release shared computational resources on demand. Moreover, it unfolds the advantage of economies of scale in the cloud through a drop in the average costs of these shared resources. However, it is still an open challenge to achieve a perfect match between resource demand and provision in autonomous elasticity management. Resource adaptation decisions essentially involve a trade-off between economics and performance, which produces a gap between the ideal and actual resource provisioning. This gap, if not properly managed, can negatively impact the aggregate utility of a cloud customer in the long run. To address this limitation, we propose a technical debt-aware learning approach for autonomous elasticity management based on a reinforcement learning of elasticity debts in resource provisioning; the adaptation pursues strategic decisions that trades off economics against performance. We extend CloudSim and Burlap to evaluate our approach. The evaluation shows that a reinforcement learning of technical debts in elasticity obtains a higher utility for a cloud customer, while conforming expected levels of performance.

Semi-simplicial spaces

This is an exposition of homotopical results on the geometric realization of semi-simplicial spaces. We then use these to derive basic foundational results about classifying spaces of topological categories, possibly without units. The topics considered include: fibrancy conditions on topological categories; the effect on classifying spaces of freely adjoining units; approximate notions of units; Quillen's Theorems A and B for non-unital topological categories; the effect on classifying spaces of changing the topology on the space of objects; the Group-Completion Theorem.

Constraints, Lazy Constraints, or Propagators in ASP Solving: An Empirical Analysis

Answer Set Programming (ASP) is a well-established declarative paradigm. One of the successes of ASP is the availability of efficient systems. State-of-the-art systems are based on the ground+solve approach. In some applications this approach is infeasible because the grounding of one or few constraints is expensive. In this paper, we systematically compare alternative strategies to avoid the instantiation of problematic constraints, that are based on custom extensions of the solver. Results on real and synthetic benchmarks highlight some strengths and weaknesses of the different strategies. (Under consideration for acceptance in TPLP, ICLP 2017 Special Issue.)

A Unified Approach to Nonlinear Transformation Materials

The advances in geometric approaches to optical devices due to transformation optics has led to the development of cloaks, concentrators, and other devices. It has also been shown that transformation optics can be used to gravitational fields from general relativity. However, the technique is currently constrained to linear devices, as a consistent approach to nonlinearity (including both the case of a nonlinear background medium and a nonlinear transformation) remains an open question. Here we show that nonlinearity can be incorporated into transformation optics in a consistent way. We use this to illustrate a number of novel effects, including cloaking an optical soliton, modeling nonlinear solutions to Einstein's field equations, controlling transport in a Debye solid, and developing a set of constitutive to relations for relativistic cloaks in arbitrary nonlinear backgrounds.

Stationary crack propagation in a two-dimensional visco-elastic network model

We investigate crack propagation in a simple two-dimensional visco-elastic model and find a scaling regime in the relation between the propagation velocity and energy release rate or fracture energy, together with lower and upper bounds of the scaling regime. On the basis of our result, the existence of the lower and upper bounds is expected to be universal or model-independent: the present simple simulation model provides generic insight into the physics of crack propagation, and the model will be a first step towards the development of a more refined coarse-grained model. Relatively abrupt changes of velocity are predicted near the lower and upper bounds for the scaling regime and the positions of the bounds could be good markers for the development of tough polymers, for which we provide simple views that could be useful as guiding principles for toughening polymer-based materials.

A note on the fundamental group of Kodaira fibrations

The fundamental group $\pi$ of a Kodaira fibration is, by definition, the extension of a surface group $\Pi_b$ by another surface group $\Pi_g$, i.e. \[ 1 \rightarrow \Pi_g \rightarrow \pi \rightarrow \Pi_b \rightarrow 1. \] Conversely, we can inquire about what conditions need to be satisfied by a group of that sort in order to be the fundamental group of a Kodaira fibration. In this short note we collect some restriction on the image of the classifying map $m \colon \Pi_b \to \Gamma_g$ in terms of the coinvariant homology of $\Pi_g$. In particular, we observe that if $\pi$ is the fundamental group of a Kodaira fibration with relative irregularity $g-s$, then $g \leq 1+ 6s$, and we show that this effectively constrains the possible choices for $\pi$, namely that there are group extensions as above that fail to satisfy this bound, hence cannot be the fundamental group of a Kodaira fibration. In particular this provides examples of symplectic $4$--manifolds that fail to admit a Kähler structure for reasons that eschew the usual obstructions.

Photo-Chemically Directed Self-Assembly of Carbon Nanotubes on Surfaces

Transistors incorporating single-wall carbon nanotubes (CNTs) as the channel material are used in a variety of electronics applications. However, a competitive CNT-based technology requires the precise placement of CNTs at predefined locations of a substrate. One promising placement approach is to use chemical recognition to bind CNTs from solution at the desired locations on a surface. Producing the chemical pattern on the substrate is challenging. Here we describe a one-step patterning approach based on a highly photosensitive surface monolayer. The monolayer contains chromophopric group as light sensitive body with heteroatoms as high quantum yield photolysis center. As deposited, the layer will bind CNTs from solution. However, when exposed to ultraviolet (UV) light with a low dose (60 mJ/cm2) similar to that used for conventional photoresists, the monolayer cleaves and no longer binds CNTs. These features allow standard, wafer-scale UV lithography processes to be used to form a patterned chemical monolayer without the need for complex substrate patterning or monolayer stamping.

Split-and-augmented Gibbs sampler - Application to large-scale inference problems

This paper derives two new optimization-driven Monte Carlo algorithms inspired from variable splitting and data augmentation. In particular, the formulation of one of the proposed approaches is closely related to the alternating direction method of multipliers (ADMM) main steps. The proposed framework enables to derive faster and more efficient sampling schemes than the current state-of-the-art methods and can embed the latter. By sampling efficiently the parameter to infer as well as the hyperparameters of the problem, the generated samples can be used to approximate Bayesian estimators of the parameters to infer. Additionally, the proposed approach brings confidence intervals at a low cost contrary to optimization methods. Simulations on two often-studied signal processing problems illustrate the performance of the two proposed samplers. All results are compared to those obtained by recent state-of-the-art optimization and MCMC algorithms used to solve these problems.

Does a generalized Chaplygin gas correctly describe the cosmological dark sector?

Yes, but only for a parameter value that makes it almost coincide with the standard model. We reconsider the cosmological dynamics of a generalized Chaplygin gas (gCg) which is split into a cold dark matter (CDM) part and a dark energy (DE) component with constant equation of state. This model, which implies a specific interaction between CDM and DE, has a $\Lambda$CDM limit and provides the basis for studying deviations from the latter. Including matter and radiation, we use the (modified) CLASS code \cite{class} to construct the CMB and matter power spectra in order to search for a gCg-based concordance model that is in agreement with the SNIa data from the JLA sample and with recent Planck data. The results reveal that the gCg parameter $\alpha$ is restricted to $|\alpha|\lesssim 0.05$, i.e., to values very close to the $\Lambda$CDM limit $\alpha =0$. This excludes, in particular, models in which DE decays linearly with the Hubble rate.

The effects of subdiffusion on the NTA size measurements of extracellular vesicles in biological samples

The interest in the extracellular vesicles (EVs) is rapidly growing as they became reliable biomarkers for many diseases. For this reason, fast and accurate techniques of EVs size characterization are the matter of utmost importance. One increasingly popular technique is the Nanoparticle Tracking Analysis (NTA), in which the diameters of EVs are calculated from their diffusion constants. The crucial assumption here is that the diffusion in NTA follows the Stokes-Einstein relation, i.e. that the Mean Square Displacement (MSD) of a particle grows linearly in time (MSD $\propto t$). However, we show that NTA violates this assumption in both artificial and biological samples, i.e. a large population of particles show a strongly sub-diffusive behaviour (MSD $\propto t^\alpha$, $0<\alpha<1$). To support this observation we present a range of experimental results for both polystyrene beads and EVs. This is also related to another problem: for the same samples there exists a huge discrepancy (by the factor of 2-4) between the sizes measured with NTA and with the direct imaging methods, such as AFM. This can be remedied by e.g. the Finite Track Length Adjustment (FTLA) method in NTA, but its applicability is limited in the biological and poly-disperse samples. On the other hand, the models of sub-diffusion rarely provide the direct relation between the size of a particle and the generalized diffusion constant. However, we solve this last problem by introducing the logarithmic model of sub-diffusion, aimed at retrieving the size data. In result, we propose a novel protocol of NTA data analysis. The accuracy of our method is on par with FTLA for small ($\simeq$200nm) particles. We apply our method to study the EVs samples and corroborate the results with AFM.

Empirical regression quantile process with possible application to risk analysis

The processes of the averaged regression quantiles and of their modifications provide useful tools in the regression models when the covariates are not fully under our control. As an application we mention the probabilistic risk assessment in the situation when the return depends on some exogenous variables. The processes enable to evaluate the expected $\alpha$-shortfall ($0\leq\alpha\leq 1$) and other measures of the risk, recently generally accepted in the financial literature, but also help to measure the risk in environment analysis and elsewhere.

Primordial perturbations from inflation with a hyperbolic field-space

We study primordial perturbations from hyperinflation, proposed recently and based on a hyperbolic field-space. In the previous work, it was shown that the field-space angular momentum supported by the negative curvature modifies the background dynamics and enhances fluctuations of the scalar fields qualitatively, assuming that the inflationary background is almost de Sitter. In this work, we confirm and extend the analysis based on the standard approach of cosmological perturbation in multi-field inflation. At the background level, to quantify the deviation from de Sitter, we introduce the slow-varying parameters and show that steep potentials, which usually can not drive inflation, can drive inflation. At the linear perturbation level, we obtain the power spectrum of primordial curvature perturbation and express the spectral tilt and running in terms of the slow-varying parameters. We show that hyperinflation with power-law type potentials has already been excluded by the recent Planck observations, while exponential-type potential with the exponent of order unity can be made consistent with observations as far as the power spectrum is concerned. We also argue that, in the context of a simple $D$-brane inflation, the hyperinflation requires exponentially large hyperbolic extra dimensions but that masses of Kaluza-Klein gravitons can be kept relatively heavy.

Role of Vanadyl Oxygen in Understanding Metallic Behavior of V2O5(001) Nanorods

Vanadium pentoxide (V2O5), the most stable member of vanadium oxide family, exhibits interesting semiconductor to metal transition in the temperature range of 530-560 K. The metallic behavior originates because of the reduction of V2O5 through oxygen vacancies. In the present report, V2O5 nanorods in the orthorhombic phase with crystal orientation of (001) are grown using vapor transport process. Among three nonequivalent oxygen atoms in a VO5 pyramidal formula unit in V2O5 structure, the role of terminal vanadyl oxygen (OI) in the formation of metallic phase above the transition temperature is established from the temperature-dependent Raman spectroscopic studies. The origin of the metallic behavior of V2O5 is also understood due to the breakdown of pdpi bond between OI and nearest V atom instigated by the formation of vanadyl OI vacancy, confirmed from the downward shift of the bottom most split-off conduction bands in the material with increasing temperature.

Graph Convolution: A High-Order and Adaptive Approach

In this paper, we presented a novel convolutional neural network framework for graph modeling, with the introduction of two new modules specially designed for graph-structured data: the $k$-th order convolution operator and the adaptive filtering module. Importantly, our framework of High-order and Adaptive Graph Convolutional Network (HA-GCN) is a general-purposed architecture that fits various applications on both node and graph centrics, as well as graph generative models. We conducted extensive experiments on demonstrating the advantages of our framework. Particularly, our HA-GCN outperforms the state-of-the-art models on node classification and molecule property prediction tasks. It also generates 32% more real molecules on the molecule generation task, both of which will significantly benefit real-world applications such as material design and drug screening.

Learning Sparse Representations in Reinforcement Learning with Sparse Coding

A variety of representation learning approaches have been investigated for reinforcement learning; much less attention, however, has been given to investigating the utility of sparse coding. Outside of reinforcement learning, sparse coding representations have been widely used, with non-convex objectives that result in discriminative representations. In this work, we develop a supervised sparse coding objective for policy evaluation. Despite the non-convexity of this objective, we prove that all local minima are global minima, making the approach amenable to simple optimization strategies. We empirically show that it is key to use a supervised objective, rather than the more straightforward unsupervised sparse coding approach. We compare the learned representations to a canonical fixed sparse representation, called tile-coding, demonstrating that the sparse coding representation outperforms a wide variety of tilecoding representations.

Almost euclidean Isoperimetric Inequalities in spaces satisfying local Ricci curvature lower bounds

Motivated by Perelman's Pseudo Locality Theorem for the Ricci flow, we prove that if a Riemannian manifold has Ricci curvature bounded below in a metric ball which moreover has almost maximal volume, then in a smaller ball (in a quantified sense) it holds an almost-euclidean isoperimetric inequality. The result is actually established in the more general framework of non-smooth spaces satisfying local Ricci curvature lower bounds in a synthetic sense via optimal transportation.

Exponential Sums and Riesz energies

We bound an exponential sum that appears in the study of irregularities of distribution (the low-frequency Fourier energy of the sum of several Dirac measures) by geometric quantities: a special case is that for all $\left\{ x_1, \dots, x_N\right\} \subset \mathbb{T}^2$, $X \geq 1$ and a universal $c>0$ $$ \sum_{i,j=1}^{N}{ \frac{X^2}{1 + X^4 \|x_i -x_j\|^4}} \lesssim \sum_{k \in \mathbb{Z}^2 \atop \|k\| \leq X}{ \left| \sum_{n=1}^{N}{ e^{2 \pi i \left\langle k, x_n \right\rangle}}\right|^2} \lesssim \sum_{i,j=1}^{N}{ X^2 e^{-c X^2\|x_i -x_j\|^2}}.$$ Since this exponential sum is intimately tied to rather subtle distribution properties of the points, we obtain nonlocal structural statements for near-minimizers of the Riesz-type energy. In the regime $X \gtrsim N^{1/2}$ both upper and lower bound match for maximally-separated point sets satisfying $\|x_i -x_j\| \gtrsim N^{-1/2}$.

One dimensionalization in the spin-1 Heisenberg model on the anisotropic triangular lattice

We investigate the effect of dimensional crossover in the ground state of the antiferromagnetic spin-$1$ Heisenberg model on the anisotropic triangular lattice that interpolates between the regime of weakly coupled Haldane chains ($J^{\prime}\! \!\ll\!\! J$) and the isotropic triangular lattice ($J^{\prime}\!\!=\!\!J$). We use the density-matrix renormalization group (DMRG) and Schwinger boson theory performed at the Gaussian correction level above the saddle-point solution. Our DMRG results show an abrupt transition between decoupled spin chains and the spirally ordered regime at $(J^{\prime}/J)_c\sim 0.42$, signaled by the sudden closing of the spin gap. Coming from the magnetically ordered side, the computation of the spin stiffness within Schwinger boson theory predicts the instability of the spiral magnetic order toward a magnetically disordered phase with one-dimensional features at $(J^{\prime}/J)_c \sim 0.43$. The agreement of these complementary methods, along with the strong difference found between the intra- and the interchain DMRG short spin-spin correlations; for sufficiently large values of the interchain coupling, suggests that the interplay between the quantum fluctuations and the dimensional crossover effects gives rise to the one-dimensionalization phenomenon in this frustrated spin-$1$ Hamiltonian.