_id
sequence
label
stringlengths
5
77
text
stringlengths
78
3.43k
[ 20 ]
math.NA cs.NA
A Deterministic Algorithm for Constructing Multiple Rank 1 Lattices of Near Optimal Size. In this paper we present the first known deterministic algorithm for the construction of multiple rank 1 lattices for the approximation of periodic functions of many variables. The algorithm works by converting a potentially large reconstructing single rank 1 lattice for some d dimensional frequency set I subset N d into a collection of much smaller rank 1 lattices which allow for accurate and efficient reconstruction of trigonometric polynomials with coefficients in I (and, therefore, for the approximation of multivariate periodic functions). The total number of sampling points in the resulting multiple rank 1 lattices is theoretically shown to be less than mathcal O left( I log 2 (N I ) right) with constants independent of d , and by performing one dimensional fast Fourier transforms on samples of trigonometric polynomials with Fourier support in I at these points, we obtain exact reconstruction of all Fourier coefficients in fewer than mathcal O left(d , I log 4 (N I ) right) total operations. Additionally, we present a second multiple rank 1 lattice construction algorithm which constructs lattices with even fewer sampling points at the cost of only being able to reconstruct exact trigonometric polynomials rather than having additional theoretical approximation. Both algorithms are tested numerically and surpass the theoretical bounds. Notably, we observe that the oversampling factors samples I appear to grow only logarithmically in I for the first algorithm and appear near optimally bounded by four in the second algorithm.
[ 40, 3 ]
cs.CL cs.IR
STEAM Self Supervised Taxonomy Expansion with Mini Paths. Taxonomies are important knowledge ontologies that underpin numerous applications on a daily basis, but many taxonomies used in practice suffer from the low coverage issue. We study the taxonomy expansion problem, which aims to expand existing taxonomies with new concept terms. We propose a self supervised taxonomy expansion model named STEAM, which leverages natural supervision in the existing taxonomy for expansion. To generate natural self supervision signals, STEAM samples mini paths from the existing taxonomy, and formulates a node attachment prediction task between anchor mini paths and query terms. To solve the node attachment task, it learns feature representations for query anchor pairs from multiple views and performs multi view co training for prediction. Extensive experiments show that STEAM outperforms state of the art methods for taxonomy expansion by 11.6 in accuracy and 7.0 in mean reciprocal rank on three public benchmarks. The implementation of STEAM can be found at url https github.com yueyu1030 STEAM .
[ 1, 2, 3 ]
cs.CL cs.AI cs.LG
BOND BERT Assisted Open Domain Named Entity Recognition with Distant Supervision. We study the open domain named entity recognition (NER) problem under distant supervision. The distant supervision, though does not require large amounts of manual annotations, yields highly incomplete and noisy distant labels via external knowledge bases. To address this challenge, we propose a new computational framework BOND, which leverages the power of pre trained language models (e.g., BERT and RoBERTa) to improve the prediction performance of NER models. Specifically, we propose a two stage training algorithm In the first stage, we adapt the pre trained language model to the NER tasks using the distant labels, which can significantly improve the recall and precision In the second stage, we drop the distant labels, and propose a self training approach to further improve the model performance. Thorough experiments on 5 benchmark datasets demonstrate the superiority of BOND over existing distantly supervised NER methods. The code and distantly labeled data have been released in https github.com cliang1453 BOND.
[ 40, 1, 3, 16 ]
cs.LG cs.CL cs.IR q-bio.QM
SumGNN Multi typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization. Thanks to the increasing availability of drug drug interactions (DDI) datasets and large biomedical knowledge graphs (KGs), accurate detection of adverse DDI using machine learning models becomes possible. However, it remains largely an open problem how to effectively utilize large and noisy biomedical KG for DDI detection. Due to its sheer size and amount of noise in KGs, it is often less beneficial to directly integrate KGs with other smaller but higher quality data (e.g., experimental data). Most of the existing approaches ignore KGs altogether. Some try to directly integrate KGs with other data via graph neural networks with limited success. Furthermore, most previous works focus on binary DDI prediction whereas the multi typed DDI pharmacological effect prediction is a more meaningful but harder task. To fill the gaps, we propose a new method SumGNN knowledge summarization graph neural network, which is enabled by a subgraph extraction module that can efficiently anchor on relevant subgraphs from a KG, a self attention based subgraph summarization scheme to generate a reasoning path within the subgraph, and a multi channel knowledge and data integration module that utilizes massive external biomedical knowledge for significantly improved multi typed DDI predictions. SumGNN outperforms the best baseline by up to 5.54 , and the performance gain is particularly significant in low data relation types. In addition, SumGNN provides interpretable prediction via the generated reasoning paths for each prediction.
[ 1, 3 ]
cs.CL cs.LG
SeqMix Augmenting Active Sequence Labeling via Sequence Mixup. Active learning is an important technique for low resource sequence labeling tasks. However, current active sequence labeling methods use the queried samples alone in each iteration, which is an inefficient way of leveraging human annotations. We propose a simple but effective data augmentation method to improve the label efficiency of active sequence labeling. Our method, SeqMix, simply augments the queried samples by generating extra labeled sequences in each iteration. The key difficulty is to generate plausible sequences along with token level labels. In SeqMix, we address this challenge by performing mixup for both sequences and token level labels of the queried samples. Furthermore, we design a discriminator during sequence mixup, which judges whether the generated sequences are plausible or not. Our experiments on Named Entity Recognition and Event Detection tasks show that SeqMix can improve the standard active sequence labeling method by 2.27 3.75 in terms of F 1 scores. The code and data for SeqMix can be found at https github.com rz zhang SeqMix
[ 1, 3 ]
cs.CL cs.LG
Fine Tuning Pre trained Language Model with Weak Supervision A Contrastive Regularized Self Training Approach. Fine tuned pre trained language models (LMs) have achieved enormous success in many natural language processing (NLP) tasks, but they still require excessive labeled data in the fine tuning stage. We study the problem of fine tuning pre trained LMs using only weak supervision, without any labeled data. This problem is challenging because the high capacity of LMs makes them prone to overfitting the noisy labels generated by weak supervision. To address this problem, we develop a contrastive self training framework, COSINE, to enable fine tuning LMs with weak supervision. Underpinned by contrastive regularization and confidence based reweighting, this contrastive self training framework can gradually improve model fitting while effectively suppressing error propagation. Experiments on sequence, token, and sentence pair classification tasks show that our model outperforms the strongest baseline by large margins on 7 benchmarks in 6 tasks, and achieves competitive performance with fully supervised fine tuning methods.
[ 64, 31 ]
physics.soc-ph q-bio.PE
Simplified model on the timing of easing the lockdown. Lockdown procedures have been proven successful in mitigating the spread of the viruses in this COVID 19 pandemic, but they also have devastating impact on the economy. We use a modified Susceptible Infectious Recovered Deceased model with time dependent infection rate to simulate how the infection is spread under lockdown. The economic cost due to the loss of workforce and incurred medical expenses is evaluated with a simple model. We find the best strategy, meaning the smallest economic cost for the entire course of the pandemic, is to keep the strict lockdown as long as possible.
[ 2, 3 ]
cs.AI cs.CL
A logic based relational learning approach to relation extraction The OntoILPER system. Relation Extraction (RE), the task of detecting and characterizing semantic relations between entities in text, has gained much importance in the last two decades, mainly in the biomedical domain. Many papers have been published on Relation Extraction using supervised machine learning techniques. Most of these techniques rely on statistical methods, such as feature based and tree kernels based methods. Such statistical learning techniques are usually based on a propositional hypothesis space for representing examples, i.e., they employ an attribute value representation of features. This kind of representation has some drawbacks, particularly in the extraction of complex relations which demand more contextual information about the involving instances, i.e., it is not able to effectively capture structural information from parse trees without loss of information. In this work, we present OntoILPER, a logic based relational learning approach to Relation Extraction that uses Inductive Logic Programming for generating extraction models in the form of symbolic extraction rules. OntoILPER takes profit of a rich relational representation of examples, which can alleviate the aforementioned drawbacks. The proposed relational approach seems to be more suitable for Relation Extraction than statistical ones for several reasons that we argue. Moreover, OntoILPER uses a domain ontology that guides the background knowledge generation process and is used for storing the extracted relation instances. The induced extraction rules were evaluated on three protein protein interaction datasets from the biomedical domain. The performance of OntoILPER extraction models was compared with other state of the art RE systems. The encouraging results seem to demonstrate the effectiveness of the proposed solution.
[ 56, 77 ]
cs.LO math.LO
Overlap Algebras a Constructive Look at Complete Boolean Algebras. The notion of a complete Boolean algebra, although completely legitimate in constructive mathematics, fails to capture some natural structures such as the lattice of subsets of a given set. Sambin's notion of an overlap algebra, although classically equivalent to that of a complete Boolean algebra, has powersets and other natural structures as instances. In this paper we study the category of overlap algebras as an extension of the category of sets and relations, and we establish some basic facts about mono epi isomorphisms and (co)limits here a morphism is a symmetrizable function (with classical logic this is just a function which preserves joins). Then we specialize to the case of morphisms which preserve also finite meets classically, this is the usual category of complete Boolean algebras. Finally, we connect overlap algebras with locales, and their morphisms with open maps between locales, thus obtaining constructive versions of some results about Boolean locales.
[ 51, 92 ]
physics.bio-ph q-bio.BM
Extended Experimental Inferential Structure Determination Method for Evaluating the Structural Ensembles of Disordered Protein States. Characterization of proteins with intrinsic or unfolded state disorder comprises a new frontier in structural biology, requiring the characterization of diverse and dynamic structural ensembles. We introduce a comprehensive Bayesian framework, the Extended Experimental Inferential Structure Determination (X EISD) method, that calculates the maximum log likelihood of a protein structural ensemble by accounting for the uncertainties of a wide range of experimental data and back calculation models from structures, including NMR chemical shifts, J couplings, Nuclear Overhauser Effects, paramagnetic relaxation enhancements, residual dipolar couplings, and hydrodynamic radii, single molecule fluorescence F "orster resonance energy transfer efficiencies and small angle X ray scattering intensity curves. We apply X EISD to the drkN SH3 unfolded state domain and show that certain experimental data types are more influential than others for both eliminating structural ensemble models, while also finding equally probable disordered ensembles that have alternative structural properties that will stimulate further experiments to discriminate between them.
[ 72, 25 ]
cs.PF math.PR
Threshold based rerouting and replication for resolving job server affinity relations. We consider a system with several job types and two parallel server pools. Within the pools the servers are homogeneous, but across pools possibly not in the sense that the service speed of a job may depend on its type as well as the server pool. Immediately upon arrival, jobs are assigned to a server pool. This could be based on (partial) knowledge of their type, but such knowledge might not be available. Information about the job type can however be obtained while the job is in service as the service progresses, the likelihood that the service speed of this job type is low increases, creating an incentive to execute the job on different, possibly faster, server(s). Two policies are considered reroute the job to the other server pool, or replicate it there. We determine the effective load per server under both the rerouting and replication policy for completely unknown as well as partly known job types. We also examine the impact of these policies on the stability bound, and find that the uncertainty in job types may significantly degrade the performance. For (highly) unbalanced service speeds full replication achieves the largest stability bound while for (nearly) balanced service speeds no replication maximizes the stability bound. Finally, we discuss how the use of threshold based policies can help improve the expected latency for completely or partly unknown job types.
[ 41, 2, 14 ]
cs.HC cs.AI cs.RO
Towards an Adaptive Robot for Sports and Rehabilitation Coaching. The work presented in this paper aims to explore how, and to what extent, an adaptive robotic coach has the potential to provide extra motivation to adhere to long term rehabilitation and help fill the coaching gap which occurs during repetitive solo practice in high performance sport. Adapting the behavior of a social robot to a specific user, using reinforcement learning (RL), could be a way of increasing adherence to an exercise routine in both domains. The requirements gathering phase is underway and is presented in this paper along with the rationale of using RL in this context.
[ 4 ]
cs.NE
Algorithm Portfolio for Individual based Surrogate Assisted Evolutionary Algorithms. Surrogate assisted evolutionary algorithms (SAEAs) are powerful optimisation tools for computationally expensive problems (CEPs). However, a randomly selected algorithm may fail in solving unknown problems due to no free lunch theorems, and it will cause more computational resource if we re run the algorithm or try other algorithms to get a much solution, which is more serious in CEPs. In this paper, we consider an algorithm portfolio for SAEAs to reduce the risk of choosing an inappropriate algorithm for CEPs. We propose two portfolio frameworks for very expensive problems in which the maximal number of fitness evaluations is only 5 times of the problem's dimension. One framework named Par IBSAEA runs all algorithm candidates in parallel and a more sophisticated framework named UCB IBSAEA employs the Upper Confidence Bound (UCB) policy from reinforcement learning to help select the most appropriate algorithm at each iteration. An effective reward definition is proposed for the UCB policy. We consider three state of the art individual based SAEAs on different problems and compare them to the portfolios built from their instances on several benchmark problems given limited computation budgets. Our experimental studies demonstrate that our proposed portfolio frameworks significantly outperform any single algorithm on the set of benchmark problems.
[ 17, 1, 15 ]
cs.IT cs.DC cs.LG math.IT
Analog Lagrange Coded Computing. A distributed computing scenario is considered, where the computational power of a set of worker nodes is used to perform a certain computation task over a dataset that is dispersed among the workers. Lagrange coded computing (LCC), proposed by Yu et al., leverages the well known Lagrange polynomial to perform polynomial evaluation of the dataset in such a scenario in an efficient parallel fashion while keeping the privacy of data amidst possible collusion of workers. This solution relies on quantizing the data into a finite field, so that Shamir's secret sharing, as one of its main building blocks, can be employed. Such a solution, however, is not properly scalable with the size of dataset, mainly due to computation overflows. To address such a critical issue, we propose a novel extension of LCC to the analog domain, referred to as analog LCC (ALCC). All the operations in the proposed ALCC protocol are done over the infinite fields of R C but for practical implementations floating point numbers are used. We characterize the privacy of data in ALCC, against any subset of colluding workers up to a certain size, in terms of the distinguishing security (DS) and the mutual information security (MIS) metrics. Also, the accuracy of outcome is characterized in a practical setting assuming operations are performed using floating point numbers. Consequently, a fundamental trade off between the accuracy of the outcome of ALCC and its privacy level is observed and is numerically evaluated. Moreover, we implement the proposed scheme to perform matrix matrix multiplication over a batch of matrices. It is observed that ALCC is superior compared to the state of the art LCC, implemented using fixed point numbers, assuming both schemes use an equal number of bits to represent data symbols.
[ 27, 13 ]
math.CT math.DS
Poly An abundant categorical setting for mode dependent dynamics. Dynamical systems by which we mean machines that take time varying input, change their state, and produce output can be wired together to form more complex systems. Previous work has shown how to allow collections of machines to reconfigure their wiring diagram dynamically, based on their collective state. This notion was called "mode dependence", and while the framework was compositional (forming an operad of re wiring diagrams and algebra of mode dependent dynamical systems on it), the formulation itself was more "creative" than it was natural. In this paper we show that the theory of mode dependent dynamical systems can be more naturally recast within the category Poly of polynomial functors. This category is almost superlatively abundant in its structure for example, it has emph four interacting monoidal structures ( , times, otimes, circ) , two of which ( times, otimes ) are monoidal closed, and the comonoids for circ are precisely categories in the usual sense. We discuss how the various structures in Poly show up in the theory of dynamical systems. We also show that the usual coalgebraic formalism for dynamical systems takes place within Poly. Indeed one can see coalgebras as special dynamical systems ones that do not record their history formally analogous to contractible groupoids as special categories.
[ 66, 6, 7, 20, 28 ]
physics.comp-ph cs.CE cs.CG cs.NA math.NA physics.flu-dyn
rp adaptation for compressible flows. We present an rp adaptation strategy for high fidelity simulation of compressible inviscid flows with shocks. The mesh resolution in regions of flow discontinuities is increased by using a variational optimiser to r adapt the mesh and cluster degrees of freedom there. In regions of smooth flow, we locally increase or decrease the local resolution through increasing or decreasing the polynomial order of the elements, respectively. This dual approach allows us to take advantage of the strengths of both methods for best computational performance, thereby reducing the overall cost of the simulation. The adaptation workflow uses a sensor for both discontinuities and smooth regions that is cheap to calculate, but the framework is general and could be used in conjunction with other feature based sensors or error estimators. We demonstrate this proof of concept using two geometries at transonic and supersonic flow regimes. The method has been implemented in the open source spectral hp element framework Nektar , and its dedicated high order mesh generation tool NekMesh . The results show that the proposed rp adaptation methodology is a reasonably cost effective way of improving accuracy.
[ 32, 17 ]
cs.DC cs.DS
Time and Space Optimal Exact Majority Population Protocols. In this paper we study population protocols governed by the em random scheduler , which uniformly at random selects pairwise interactions between n agents. The main result of this paper is the first time and space optimal em exact majority population protocol which also works with high probability. The new protocol operates in the optimal em parallel time O( log n), which is equivalent to O(n log n) sequential em pairwise interactions , where each agent utilises the optimal number of O( log n) states. The time optimality of the new majority protocol is possible thanks to the novel concept of fixed resolution phase clocks introduced and analysed in this paper. The new phase clock allows to count approximately constant parallel time in population protocols.
[ 6 ]
physics.flu-dyn
Theoretical framework to surpass the Betz limit using unsteady fluid mechanics. The Betz limit expresses the maximum proportion of the kinetic energy flux incident on an energy conversion device that can be extracted from an unbounded flow. The derivation of the Betz limit requires an assumption of steady flow through a notional actuator disk that is stationary in the streamwise direction. The present derivation relaxes the assumptions of steady flow and streamwise actuator disk stationarity, which expands the physically realizable parameter space of flow conditions upstream and downstream of the actuator disk. A key consequence of this generalization is the existence of unsteady motions that can, in principle, lead to energy conversion efficiencies that exceed the Betz limit not only transiently, but also in time averaged performance. Potential physical implementations of those unsteady motions are speculated.
[ 40, 1, 3 ]
cs.IR cs.CL cs.LG
Shallow Domain Adaptive Embeddings for Sentiment Analysis. This paper proposes a way to improve the performance of existing algorithms for text classification in domains with strong language semantics. We propose a domain adaptation layer learns weights to combine a generic and a domain specific (DS) word embedding into a domain adapted (DA) embedding. The DA word embeddings are then used as inputs to a generic encoder classifier framework to perform a downstream task such as classification. This adaptation layer is particularly suited to datasets that are modest in size, and which are, therefore, not ideal candidates for (re)training a deep neural network architecture. Results on binary and multi class classification tasks using popular encoder architectures, including current state of the art methods (with and without the shallow adaptation layer) show the effectiveness of the proposed approach.
[ 64, 59, 39 ]
physics.soc-ph cs.CY cs.SI
The wisdom of the few Predicting collective success from individual behavior. Can we predict top performing products, services, or businesses by only monitoring the behavior of a small set of individuals? Although most previous studies focused on the predictive power of "hub" individuals with many social contacts, which sources of customer behavioral data are needed to address this question remains unclear, mostly due to the scarcity of available datasets that simultaneously capture individuals' purchasing patterns and social interactions. Here, we address this question in a unique, large scale dataset that combines individuals' credit card purchasing history with their social and mobility traits across an entire nation. Surprisingly, we find that the purchasing history alone enables the detection of small sets of discoverers" whose early purchases offer reliable success predictions for the brick and mortar stores they visit. In contrast with the assumptions by most existing studies on word of mouth processes, the hubs selected by social network centrality are not consistently predictive of success. Our findings show that companies and organizations with access to large scale purchasing data can detect the discoverers and leverage their behavior to anticipate market trends, without the need for social network data.
[ 26, 21 ]
math.CO math.RT
A combinatorial expansion of vertical strip LLT polynomials in the basis of elementary symmetric functions. We give a new characterization of the vertical strip LLT polynomials mathrm LLT P(x q) as the unique family of symmetric functions that satisfy certain combinatorial relations. This characterization is then used to prove an explicit combinatorial expansion of vertical strip LLT polynomials in terms of elementary symmetric functions. Such formulas were conjectured independently by A. Garsia et al. and the first named author, and are governed by the combinatorics of orientations of unit interval graphs. The obtained expansion is manifestly positive if q is replaced by q 1 , thus recovering a recent result of M. D'Adderio. Our results are based on linear relations among LLT polynomials that arise in the work of D'Adderio, and of E. Carlsson and A. Mellit. To some extent these relations are given new bijective proofs using colorings of unit interval graphs. As a bonus we obtain a new characterization of chromatic quasisymmetric functions of unit interval graphs.
[ 25, 54 ]
math.PR q-fin.MF
Vol of vol expansion for (rough) stochastic volatility models. We introduce an asymptotic small noise expansion, a so called vol of vol expansion, for potentially infinite dimensional and rough stochastic volatility models. Thereby we extend the scope of existing results for finite dimensional models and validate claims for infinite dimensional models. Furthermore we provide new, explicit (in the sense of non recursive) representations of the so called push down Malliavin weights that utilizes a precise understanding of the terms of this expansion.
[ 46, 15 ]
cs.IT math.AG math.IT
A general family of MSRD codes and PMDS codes with smaller field sizes from extended Moore matrices. We construct six new explicit families of linear maximum sum rank distance (MSRD) codes, each of which has the smallest field sizes among all known MSRD codes for some parameter regime. Using them and a previous result of the author, we provide two new explicit families of linear partial MDS (PMDS) codes with smaller field sizes than previous PMDS codes for some parameter regimes. Our approach is to characterize evaluation points that turn extended Moore matrices into the parity check matrix of a linear MSRD code. We then produce such sequences from codes with good Hamming metric parameters. The six new families of linear MSRD codes with smaller field sizes are obtained using MDS codes, Hamming codes, BCH codes and three Algebraic Geometry codes. The MSRD codes based on Hamming codes, of minimum sum rank distance 3 , meet a recent bound by Byrne et al.
[ 64, 11 ]
physics.soc-ph physics.app-ph
Discontinuous transition to loop formation in optimal supply networks. The structure and design of optimal supply networks is an important topic in complex networks research. A fundamental trait of natural and man made networks is the emergence of loops and the trade off governing their formation adding redundant edges to supply networks is costly, yet beneficial for resilience. Loops typically form when costs for new edges are small or inputs uncertain. Here, we shed further light on the transition to loop formation. We demonstrate that loops emerge discontinuously when decreasing the costs for new edges for both an edge damage model and a fluctuating sink model. Mathematically, new loops are shown to form through a saddle node bifurcation. Our analysis allows to heuristically predict the location and cost where the first loop emerges. Finally, we unveil an intimate relationship among betweenness measures and optimal tree networks. Our results can be used to understand the evolution of loop formation in real world biological networks.
[ 17, 43 ]
cs.DB cs.DC
PnyxDB a Lightweight Leaderless Democratic Byzantine Fault Tolerant Replicated Datastore. Byzantine Fault Tolerant (BFT) systems are rapidly emerging as a viable technology for production grade systems, notably in closed consortia deployments for nancial and supply chain applications. Unfortunately, most algorithms proposed so far to coordinate these systems suffer from substantial scalability issues, and lack important features to implement Internet scale governance mechanisms. In this paper, we observe that many application workloads offer little concurrency, and propose PnyxDB, an eventually consistent Byzantine Fault Tolerant replicated datastore that exhibits both high scalability and low latency. Our approach is based on conditional endorsements, that allow nodes to specify the set of transactions that must not be committed for the endorsement to be valid. In addition to its high scalability, PnyxDB supports application level voting, i.e. individual nodes are able to endorse or reject a transaction according to application defined policies without compromising consistency. We provide a comparison against BFTSMaRt and Tendermint, two competitors with different design aims, and show that our implementation speeds up commit latencies by a factor of 11, remaining below 5 seconds in a worldwide geodistributed deployment of 180 nodes.
[ 49 ]
math.OC
The inf translation for solving set minimization problems. Set and vector valued optimization problems can be re formulated as complete lattice valued problems. This has several advantages, one of which is the existence of a clear cut solution concept which includes the attainment as the infimum (not present in traditional vector optimization theory) and minimality as two potentially different features. The task is to find a set which is large enough to generate the infimum but at the same time small enough to include only minimizers. In this paper, optimality conditions for such sets based on the inf translation are given within an abstract framework. The inf translation reduces the solution set to a single point which in turn admits the application of more standard procedures. For functions with values in complete lattices of sets, scalarization results are provided where the focus is on convex problems. Vector optimization problems, in particular a vectorial calculus of variations problem, are discussed as examples.
[ 46 ]
math.AG
Euclidean distance degree and mixed volume. We initiate a study of the Euclidean Distance Degree in the context of sparse polynomials. Specifically, we consider a hypersurface f 0 defined by a polynomial f that is general given its support, such that the support contains the origin. We show that the Euclidean Distance Degree of f 0 equals the mixed volume of the Newton polytopes of the associated Lagrange multiplier equations. We discuss the implication of our result for computational complexity and give a formula for the Euclidean distance degree when the Newton polytope is a rectangular parallelepiped.
[ 17 ]
cs.DC
The Scalable Systems Laboratory a Platform for Software Innovation for HEP. The Scalable Systems Laboratory (SSL), part of the IRIS HEP Software Institute, provides Institute participants and HEP software developers generally with a means to transition their R D from conceptual toys to testbeds to production scale prototypes. The SSL enables tooling, infrastructure, and services supporting the innovation of novel analysis and data architectures, development of software elements and tool chains, reproducible functional and scalability testing of service components, and foundational systems R D for accelerated services developed by the Institute. The SSL is constructed with a core team having expertise in scale testing and deployment of services across a wide range of cyberinfrastructure. The core team embeds and partners with other areas in the Institute, and with LHC and other HEP development and operations teams as appropriate, to define investigations and required service deployment patterns. We describe the approach and experiences with early application deployments, including analysis platforms and intelligent data delivery systems.
[ 18, 29 ]
math.GT math.GR
Characterizing divergence and thickness in right angled Coxeter groups. We completely classify the possible divergence functions for right angled Coxeter groups (RACGs). In particular, we show that the divergence of any such group is either polynomial, exponential or infinite. We prove that a RACG is strongly thick of order k if and only if its divergence function is a polynomial of degree k 1. Moreover, we show that the exact divergence function of a RACG can easily be computed from its defining graph by an invariant we call the hypergraph index.
[ 49 ]
math.OC
Estimator Based Output Feedback Stabilization of Linear Multi Delay Systems using SOS. In this paper, we investigate the estimator based output feedback control problem of multi delay systems. This work is an extension of recently developed operator value LMI framework for infinite dimensional time delay systems. Based on the optimal convex state feedback controller and generalized Luenberger observer synthesis conditions we already have, the estimator based output feedback controller is designed to contain the estimates of both the present state and history of the state. An output feedback controller synthesis condition is proposed using SOS method, which is expressed in a set of LMI SDP constraints. The simulation examples are displayed to demonstrate the effectiveness and advantages of the proposed results.
[ 1, 2, 49 ]
cs.LG cs.AI math.OC
Learning to Optimise General TSP Instances. The Travelling Salesman Problem (TSP) is a classical combinatorial optimisation problem. Deep learning has been successfully extended to meta learning, where previous solving efforts assist in learning how to optimise future optimisation instances. In recent years, learning to optimise approaches have shown success in solving TSP problems. However, they focus on one type of TSP problem, namely ones where the points are uniformly distributed in Euclidean spaces and have issues in generalising to other embedding spaces, e.g., spherical distance spaces, and to TSP instances where the points are distributed in a non uniform manner. An aim of learning to optimise is to train once and solve across a broad spectrum of (TSP) problems. Although supervised learning approaches have shown to achieve more optimal solutions than unsupervised approaches, they do require the generation of training data and running a solver to obtain solutions to learn from, which can be time consuming and difficult to find reasonable solutions for harder TSP instances. Hence this paper introduces a new learning based approach to solve a variety of different and common TSP problems that are trained on easier instances which are faster to train and are easier to obtain better solutions. We name this approach the non Euclidean TSP network (NETSP Net). The approach is evaluated on various TSP instances using the benchmark TSPLIB dataset and popular instance generator used in the literature. We performed extensive experiments that indicate our approach generalises across many types of instances and scales to instances that are larger than what was used during training.
[ 74 ]
math.OA
C extreme points of positive operator valued measures and unital completely positive maps. We study the quantum ( C ) convexity structure of normalized positive operator valued measures (POVMs) on measurable spaces. In particular, it is seen that unlike extreme points under classical convexity, C extreme points of normalized POVMs on countable spaces (in particular for finite sets) are always spectral measures (normalized projection valued measures). More generally it is shown that atomic C extreme points are spectral. A Krein Milman type theorem for POVMs has also been proved. As an application it is shown that a map on any commutative unital C algebra with countable spectrum (in particular mathbb C n ) is C extreme in the set of unital completely positive maps if and only if it is a unital homomorphism.
[ 13 ]
math.DS
Discovering mean residence time and escape probability from data of stochastic dynamical systems. We present a method to learn mean residence time and escape probability from data modeled by stochastic differential equations. This method is a combination of machine learning from data (to extract stochastic differential equations as models) and stochastic dynamics (to quantify dynamical behaviors with deterministic tools). The goal is to learn and understand stochastic dynamics based on data. This method is applicable to sample path data collected from complex systems, as long as these systems can be modeled as stochastic differential equations.
[ 9, 20 ]
math.NA cs.NA math.AP
Optimal operator preconditioning for pseudodifferential boundary problems. We propose an operator preconditioner for general elliptic pseudodifferential equations in a domain Omega , where Omega is either in mathbb R n or in a Riemannian manifold. For linear systems of equations arising from low order Galerkin discretizations, we obtain condition numbers that are independent of the mesh size and of the choice of bases for test and trial functions. The basic ingredient is a classical formula by Boggio for the fractional Laplacian, which is extended analytically. In the special case of the weakly and hypersingular operators on a line segment or a screen, our approach gives a unified, independent proof for a series of recent results by Hiptmair, Jerez Hanckes, N ' e d ' e lec and Urz ' u a Torres. We also study the increasing relevance of the regularity assumptions on the mesh with the order of the operator. Numerical examples validate our theoretical findings and illustrate the performance of the proposed preconditioner on quasi uniform, graded and adaptively generated meshes.
[ 1, 2, 3 ]
cs.CL cs.AI cs.LG
Emotion Correlation Mining Through Deep Learning Models on Natural Language Text. Emotion analysis has been attracting researchers' attention. Most previous works in the artificial intelligence field focus on recognizing emotion rather than mining the reason why emotions are not or wrongly recognized. Correlation among emotions contributes to the failure of emotion recognition. In this paper, we try to fill the gap between emotion recognition and emotion correlation mining through natural language text from web news. Correlation among emotions, expressed as the confusion and evolution of emotion, is primarily caused by human emotion cognitive bias. To mine emotion correlation from emotion recognition through text, three kinds of features and two deep neural network models are presented. The emotion confusion law is extracted through orthogonal basis. The emotion evolution law is evaluated from three perspectives, one step shift, limited step shifts, and shortest path transfer. The method is validated using three datasets the titles, the bodies, and the comments of news articles, covering both objective and subjective texts in varying lengths (long and short). The experimental results show that, in subjective comments, emotions are easily mistaken as anger. Comments tend to arouse emotion circulations of love anger and sadness anger. In objective news, it is easy to recognize text emotion as love and cause fear joy circulation. That means, journalists may try to attract attention using fear and joy words but arouse the emotion love instead After news release, netizens generate emotional comments to express their intense emotions, i.e., anger, sadness, and love. These findings could provide insights for applications regarding affective interaction such as network public sentiment, social media communication, and human computer interaction.
[ 0 ]
cs.CV
Collaborative Attention Mechanism for Multi View Action Recognition. Multi view action recognition (MVAR) leverages complementary temporal information from different views to improve the learning performance. Obtaining informative view specific representation plays an essential role in MVAR. Attention has been widely adopted as an effective strategy for discovering discriminative cues underlying temporal data. However, most existing MVAR methods only utilize attention to extract representation for each view individually, ignoring the potential to dig latent patterns based on mutual support information in attention space. To this end, we propose a collaborative attention mechanism (CAM) for solving the MVAR problem in this paper. The proposed CAM detects the attention differences among multi view, and adaptively integrates frame level information to benefit each other. Specifically, we extend the long short term memory (LSTM) to a Mutual Aid RNN (MAR) to achieve the multi view collaboration process. CAM takes advantages of view specific attention pattern to guide another view and discover potential information which is hard to be explored by itself. It paves a novel way to leverage attention information and enhances the multi view representation learning. Extensive experiments on four action datasets illustrate the proposed CAM achieves better results for each view and also boosts multi view performance.
[ 3 ]
cs.CL
A Comprehensive Comparison of Machine Learning Based Methods Used in Bengali Question Classification. QA classification system maps questions asked by humans to an appropriate answer category. A sound question classification (QC) system model is the pre requisite of a sound QA system. This work demonstrates phases of assembling a QA type classification model. We present a comprehensive comparison (performance and computational complexity) among some machine learning based approaches used in QC for Bengali language.
[ 11 ]
physics.app-ph
Perfect anomalous splitter by acoustic meta grating. As an inversely designed artificial device, metasurface usually means densely arranged meta atoms with complex substructures. In acoustics, those meta atoms are usually constructed by multi folded channels or multi connected cavities of deep sub wavelength feature, which limits their implementation in pragmatic applications. We propose here a comprehensive concept of a perfect anomalous splitter based on an acoustic meta grating. The beam splitter is designed by etching only two or four straight walled grooves per period on a planar hard surface. Different from the recently reported reflectors or splitters, our device can perfectly split an incident wave into different desired directions with arbitrary power flow partition. In addition, because ultrathin substructures with thin walls and narrow channels are avoided in our design procedure, the proposed beam splitter can be used for waves with much shorter wavelength compared to the previous suggested systems. The design is established by rigorous formulae developed under the framework of the grating theory and a genetic optimization algorithm. Numerical simulation and experimental evidence are provided to discuss the involved physical mechanism and to give the proof of concept for the proposed perfect anomalous acoustic splitter.
[ 9 ]
math.AP
On the Cauchy problem for the pressureless Euler Navier Stokes system in the whole space. In this paper, we study the global Cauchy problem for a two phase fluid model consisting of the pressureless Euler equations and the incompressible Navier Stokes equations where the coupling of two equations is through the drag force. We establish the global in time existence and uniqueness of classical solutions for that system when the initial data are sufficiently small and regular. Main difficulties arise in the absence of pressure in the Euler equations. In order to resolve it, we properly combine the large time behavior of classical solutions and the bootstrapping argument to construct the global in time unique classical solutions.
[ 38 ]
math.NT
Lagrange spectrum of a circle over the Eisensteinian field. We study an intrinsic Lagrange spectrum of the unit circle z 1 in the complex plane with respect to the Eisensteinian field mathbb Q ( sqrt 3 ) . We prove that the minimum of the Lagrange spectrum is 2 and that its smallest accumulation point is 4 sqrt 3 . In addition, we characterize the set of all values in the spectrum between 2 and 4 sqrt3 .
[ 1, 39 ]
cs.CY cs.LG
An Ambient Physical System to Infer Concentration in Open plan Workplace. One of the core challenges in open plan workspaces is to ensure a good level of concentration for the workers while performing their tasks. Hence, being able to infer concentration levels of workers will allow building designers, managers, and workers to estimate what effect different open plan layouts will have and to find an optimal one. In this research, we present an ambient physical system to investigate the concentration inference problem. Specifically, we deploy a series of pervasive sensors to capture various ambient and physical signals related to perceived concentration at work. The practicality of our system has been tested on two large open plan workplaces with different designs and layouts. The empirical results highlight promising applications of pervasive sensing in occupational concentration inference, which can be adopted to enhance the capabilities of modern workplaces.
[ 20 ]
math.NA cs.NA
Method of fundamental solutions for Neumann problems of the modified Helmholtz equation in disk domains. The method of the fundamental solutions (MFS) is used to construct an approximate solution for a partial differential equation in a bounded domain. It is demonstrated by combining the fundamental solutions shifted to the points outside the domain and determining the coefficients of the linear sum to satisfy the boundary condition on the finite points of the boundary. In this paper, the existence of the approximate solution by the MFS for the Neumann problems of the modified Helmholtz equation in disk domains is rigorously demonstrated. We reveal the sufficient condition of the existence of the approximate solution. Applying Green's theorem to the Neumann problem of the modified Helmholtz equation, we bound the error between the approximate solution and exact solution into the difference of the function of the boundary condition and the normal derivative of the approximate solution by boundary integrations. Using this estimate of the error, we show the convergence of the approximate solution by the MFS to the exact solution with exponential order, that is, N 2a N order, where a is a positive constant less than one and N is the number of collocation points. Furthermore, it is demonstrated that the error tends to 0 in exponential order in the numerical simulations with increasing number of collocation points N .
[ 71 ]
math.SG
Special eccentricities of rational four dimensional ellipsoids. A striking result of McDuff and Schlenk asserts that in determining when a four dimensional symplectic ellipsoid can be symplectically embedded into a four dimensional symplectic ball, the answer is governed by an "infinite staircase" determined by the odd index Fibonacci numbers and the Golden Mean. Here we study embeddings of one four dimensional symplectic ellipsoid into another, and we show that if the target is rational, then the infinite staircase phenomenon found by McDuff and Schlenk is quite rare. Specifically, in the rational case, there is an infinite staircase in precisely three cases when the target has "eccentricity" 1, 2, or 3 2 in all other cases the answer is given by the classical volume obstruction except on finitely many compact intervals on which it is linear. This verifies in the special case of ellipsoids a conjecture by Holm, Mandini, Pires, and the author.
[ 38 ]
math.NT
On the unit equation over cyclic number fields of prime degree. Let ell ne 3 be a prime. We show that there are only finitely many cyclic number fields F of degree ell for which the unit equation lambda mu 1, qquad lambda, mu in mathcal O F times has solutions. Our result is effective. For example, we deduce that the only cyclic quintic number field for which the unit equation has solutions is mathbb Q ( zeta 11 ) .
[ 2, 43 ]
cs.DB cs.AI
A Toolkit for Generating Code Knowledge Graphs. Knowledge graphs have been proven extremely useful in powering diverse applications in semantic search and natural language understanding. In this paper, we present GraphGen4Code, a toolkit to build code knowledge graphs that can similarly power various applications such as program search, code understanding, bug detection, and code automation. GraphGen4Code uses generic techniques to capture code semantics with the key nodes in the graph representing classes, functions, and methods. Edges indicate function usage (e.g., how data flows through function calls, as derived from program analysis of real code), and documentation about functions (e.g., code documentation, usage documentation, or forum discussions such as StackOverflow). Our toolkit uses named graphs in RDF to model graphs per program, or can output graphs as JSON. We show the scalability of the toolkit by applying it to 1.3 million Python files drawn from GitHub, 2,300 Python modules, and 47 million forum posts. This results in an integrated code graph with over 2 billion triples. We make the toolkit to build such graphs as well as the sample extraction of the 2 billion triples graph publicly available to the community for use.
[ 27, 47 ]
math.AT math.CT
An extension of Brown functor to cospans of spaces. Let mathcal A be a small abelian category. The purpose of this paper is to introduce and study a category overline mathcal A which implicitly appears in construction of some TQFTs where overline mathcal A is determined by mathcal A . If mathcal A is the category of abelian groups, then the TQFTs obtained by Dijkgraaf Witten theory of abelian groups or Turaev Viro theory of bicommutative Hopf algebras factor through overline mathcal A up to a scaling. In this paper, we go further by giving a sufficient condition for an mathcal A valued Brown functor to extend to a homotopy theoretic analogue of overline mathcal A valued TQFT for arbitrary mathcal A . The results of this paper and our subsequent paper reproduces TQFTs obtained by DW theory and TV theory.
[ 20 ]
math.NA cs.NA
A projection based Variational Multiscale Method for Atmosphere Ocean Interaction. The proposed method aims to approximate a solution of a fluid fluid interaction problem in case of low viscosities. The nonlinear interface condition on the joint boundary allows for this problem to be viewed as a simplified version of the atmosphere ocean coupling. Thus, the proposed method should be viewed as potentially applicable to air sea coupled flows in turbulent regime. The method consists of two key ingredients. The geometric averaging approach is used for efficient and stable decoupling of the problem, which would allow for the usage of preexisting codes for the air and sea domain separately, as "black boxes". This is combined with the variational multiscale stabilization technique for treating flows at high Reynolds numbers. We prove the stability and accuracy of the method and provide several numerical tests to assess both the quantitative and qualitative features of the computed solution.
[ 38, 21, 70 ]
math.HO math.CO math.NT
Generalized Rascal Triangles. The Rascal Triangle was introduced by three middle school students in 2010, and in this paper we describe number triangles that are generalizations of the Rascal Triangle and show that these Generalized Rascal Triangles are characterized by arithmetic sequences on all diagonals as well as a Rascal like multiplication and addition rules.
[ 17 ]
cs.DC
Front Running Protection for Distributed Exchanges using Tamper Resistant Round Trip Time Measurements. In this paper we present ODIN, a front running protection system that uses a novel algorithm to measure Round Trip Time (RTT) to untrusted servers. ODIN is the decentralized equivalent of THOR, a RTT aware front running protection system for trading on centralized exchanges. Unlike centralized exchanges, P2P exchanges have potentially malicious peers which makes reliable direct RTT measurement impossible. In order to prevent tampering by an arbitrarily malicious peer, ODIN performs an indirect RTT measurement that never interacts directly with the target machine. The RTT to the target is estimated by measuring the RTT to a randomized IP address that is known to be close to the target's IP address in the global routing network. We find that ODIN's RTT estimation algorithm provides an accurate, practical, and generic solution for collecting network latency data in a hostile network environment.
[ 13 ]
math.DS
Uniform recurrence properties for beta transformation. For any beta 1 , let T beta 0,1) rightarrow 0,1) be the beta transformation defined by T beta x beta x mod 1 . We study the uniform recurrence properties of the orbit of a point under the beta transformation to the point itself. The size of the set of points with prescribed uniform recurrence rate is obtained. More precisely, for any 0 leq hat r leq infty , the set left x in 0,1) forall N gg1, exists 1 leq n leq N, rm s.t. T n beta x x leq beta hat r N right is of Hausdorff dimension left( frac 1 hat r 1 hat r right) 2 if 0 leq hat r leq 1 and is countable if hat r 1 .
[ 2 ]
cs.AI
Do We Really Sample Right In Model Based Diagnosis?. Statistical samples, in order to be representative, have to be drawn from a population in a random and unbiased way. Nevertheless, it is common practice in the field of model based diagnosis to make estimations from (biased) best first samples. One example is the computation of a few most probable possible fault explanations for a defective system and the use of these to assess which aspect of the system, if measured, would bring the highest information gain. In this work, we scrutinize whether these statistically not well founded conventions, that both diagnosis researchers and practitioners have adhered to for decades, are indeed reasonable. To this end, we empirically analyze various sampling methods that generate fault explanations. We study the representativeness of the produced samples in terms of their estimations about fault explanations and how well they guide diagnostic decisions, and we investigate the impact of sample size, the optimal trade off between sampling efficiency and effectivity, and how approximate sampling techniques compare to exact ones.
[ 81, 76 ]
cs.MS math.SP
Randomized Projection for Rank Revealing Matrix Factorizations and Low Rank Approximations. Rank revealing matrix decompositions provide an essential tool in spectral analysis of matrices, including the Singular Value Decomposition (SVD) and related low rank approximation techniques. QR with Column Pivoting (QRCP) is usually suitable for these purposes, but it can be much slower than the unpivoted QR algorithm. For large matrices, the difference in performance is due to increased communication between the processor and slow memory, which QRCP needs in order to choose pivots during decomposition. Our main algorithm, Randomized QR with Column Pivoting (RQRCP), uses randomized projection to make pivot decisions from a much smaller sample matrix, which we can construct to reside in a faster level of memory than the original matrix. This technique may be understood as trading vastly reduced communication for a controlled increase in uncertainty during the decision process. For rank revealing purposes, the selection mechanism in RQRCP produces results that are the same quality as the standard algorithm, but with performance near that of unpivoted QR (often an order of magnitude faster for large matrices). We also propose two formulas that facilitate further performance improvements. The first efficiently updates sample matrices to avoid computing new randomized projections. The second avoids large trailing updates during the decomposition in truncated low rank approximations. Our truncated version of RQRCP also provides a key initial step in our truncated SVD approximation, TUXV. These advances open up a new performance domain for large matrix factorizations that will support efficient problem solving techniques for challenging applications in science, engineering, and data analysis.
[ 60 ]
cs.AR
hXDP Efficient Software Packet Processing on FPGA NICs. FPGA accelerators on the NIC enable the offloading of expensive packet processing tasks from the CPU. However, FPGAs have limited resources that may need to be shared among diverse applications, and programming them is difficult. We present a solution to run Linux's eXpress Data Path programs written in eBPF on FPGAs, using only a fraction of the available hardware resources while matching the performance of high end CPUs. The iterative execution model of eBPF is not a good fit for FPGA accelerators. Nonetheless, we show that many of the instructions of an eBPF program can be compressed, parallelized or completely removed, when targeting a purpose built FPGA executor, thereby significantly improving performance. We leverage that to design hXDP, which includes (i) an optimizing compiler that parallelizes and translates eBPF bytecode to an extended eBPF Instruction set Architecture defined by us a (ii) soft CPU to execute such instructions on FPGA and (iii) an FPGA based infrastructure to provide XDP's maps and helper functions as defined within the Linux kernel. We implement hXDP on an FPGA NIC and evaluate it running real world unmodified eBPF programs. Our implementation is clocked at 156.25MHz, uses about 15 of the FPGA resources, and can run dynamically loaded programs. Despite these modest requirements, it achieves the packet processing throughput of a high end CPU core and provides a 10x lower packet forwarding latency.
[ 64 ]
physics.soc-ph
Incentive driven transition to high ride sharing adoption. Ride sharing the combination of multiple trips into one may substantially contribute towards sustainable urban mobility. It is most efficient at high demand locations with many similar trip requests. However, here we reveal that people's willingness to share rides does not follow this trend. Modeling the fundamental incentives underlying individual ride sharing decisions, we find two opposing adoption regimes, one with constant and another one with decreasing adoption as demand increases. In the high demand limit, the transition between these regimes becomes discontinuous, switching abruptly from low to high ride sharing adoption. Analyzing over 360 million ride requests in New York City and Chicago illustrates that both regimes coexist across the cities, consistent with our model predictions. These results suggest that even a moderate increase in the financial incentives may have a disproportionately large effect on the ride sharing adoption of individual user groups.
[ 0, 1 ]
cs.CV cs.LG
PyTorch Metric Learning. Deep metric learning algorithms have a wide variety of applications, but implementing these algorithms can be tedious and time consuming. PyTorch Metric Learning is an open source library that aims to remove this barrier for both researchers and practitioners. The modular and flexible design allows users to easily try out different combinations of algorithms in their existing code. It also comes with complete train test workflows, for users who want results fast. Code and documentation is available at https www.github.com KevinMusgrave pytorch metric learning.
[ 1, 3 ]
cs.LG cs.CL
A Probabilistic Generative Model for Typographical Analysis of Early Modern Printing. We propose a deep and interpretable probabilistic generative model to analyze glyph shapes in printed Early Modern documents. We focus on clustering extracted glyph images into underlying templates in the presence of multiple confounding sources of variance. Our approach introduces a neural editor model that first generates well understood printing phenomena like spatial perturbations from template parameters via interpertable latent variables, and then modifies the result by generating a non interpretable latent vector responsible for inking variations, jitter, noise from the archiving process, and other unforeseen phenomena associated with Early Modern printing. Critically, by introducing an inference network whose input is restricted to the visual residual between the observation and the interpretably modified template, we are able to control and isolate what the vector valued latent variable captures. We show that our approach outperforms rigid interpretable clustering baselines (Ocular) and overly flexible deep generative models (VAE) alike on the task of completely unsupervised discovery of typefaces in mixed font documents.
[ 21 ]
math.CO
Inequalities of Independence Number, Clique Number and Connectivity of Maximal Connected Domination Critical Graphs. A k gamma c edge critical graph is a graph G with the connected domination number gamma c (G) k and gamma c (G uv) k for every uv in E( overline G ) . Further, a 2 connected graph G is said to be k gamma c vertex critical if gamma c (G) k and gamma c (G v) k for all v in V(G) . A maximal k gamma c vertex critical graph is a graph which are both k gamma c edge critical and k gamma c vertex critical. Let kappa, delta, omega and alpha be respectively connectivity minimum degree, clique number and independence number. In this paper, we prove that every maximal 3 gamma c vertex critical graph G satisfies alpha leq delta and this bound is best possible. We prove further that G satisfies alpha omega leq n 1 and we also characterize all such graphs achieving the upper bounds. We finally show that if G satisfies kappa delta , then every two vertices of G are joined by hamiltonian path.
[ 24, 42 ]
math.MG math.DG
Locally homogeneous C 0 Riemannian manifolds. We show that locally homogeneous mathcal C 0 Riemannian manifolds are smooth.
[ 21 ]
math.CO
Packing branchings under cardinality constraints on their root sets. Edmonds' fundamental theorem on arborescences characterizes the existence of k pairwise arc disjoint spanning arborescences with prescribed root sets in a digraph. In this paper, we study the problem of packing branchings in digraphs under cardinality constraints on their root sets by arborescence augmentation. Let D (V x,A) be a digraph, mathcal P I 1 , ldots, I l be a partition of k , c 1 , ldots, c l , c' 1 , ldots, c' l be nonnegative integers such that c alpha leq c' alpha for alpha in l , F 1 , ldots, F k be k arc disjoint x arborescences in D such that sum i in I alpha d F i (x) leq c' alpha for alpha in l . We give a characterization on when F 1 , ldots, F k can be completed to arc disjoint spanning x arborescences F 1 , ldots, F k such that for any alpha in l , c alpha leq sum i in I alpha d F i (x) leq c' alpha .
[ 0, 41 ]
cs.RO cs.CV
Good Feature Matching Towards Accurate, Robust VO VSLAM with Low Latency. Analysis of state of the art VO VSLAM system exposes a gap in balancing performance (accuracy robustness) and efficiency (latency). Feature based systems exhibit good performance, yet have higher latency due to explicit data association direct semidirect systems have lower latency, but are inapplicable in some target scenarios or exhibit lower accuracy than feature based ones. This paper aims to fill the performance efficiency gap with an enhancement applied to feature based VSLAM. We present good feature matching, an active map to frame feature matching method. Feature matching effort is tied to submatrix selection, which has combinatorial time complexity and requires choosing a scoring metric. Via simulation, the Max logDet matrix revealing metric is shown to perform best. For real time applicability, the combination of deterministic selection and randomized acceleration is studied. The proposed algorithm is integrated into monocular stereo feature based VSLAM systems. Extensive evaluations on multiple benchmarks and compute hardware quantify the latency reduction and the accuracy robustness preservation.
[ 0 ]
cs.CV
Joint Learning of Self Representation and Indicator for Multi View Image Clustering. Multi view subspace clustering aims to divide a set of multisource data into several groups according to their underlying subspace structure. Although the spectral clustering based methods achieve promotion in multi view clustering, their utility is limited by the separate learning manner in which affinity matrix construction and cluster indicator estimation are isolated. In this paper, we propose to jointly learn the self representation, continue and discrete cluster indicators in an unified model. Our model can explore the subspace structure of each view and fusion them to facilitate clustering simultaneously. Experimental results on two benchmark datasets demonstrate that our method outperforms other existing competitive multi view clustering methods.
[ 3 ]
cs.CL
Model based annotation of coreference. Humans do not make inferences over texts, but over models of what texts are about. When annotators are asked to annotate coreferent spans of text, it is therefore a somewhat unnatural task. This paper presents an alternative in which we preprocess documents, linking entities to a knowledge base, and turn the coreference annotation task in our case limited to pronouns into an annotation task where annotators are asked to assign pronouns to entities. Model based annotation is shown to lead to faster annotation and higher inter annotator agreement, and we argue that it also opens up for an alternative approach to coreference resolution. We present two new coreference benchmark datasets, for English Wikipedia and English teacher student dialogues, and evaluate state of the art coreference resolvers on them.
[ 26 ]
math.RT
The ell modular representation of reductive groups over finite local rings of length two. Let mathcal O 2 and mathcal O ' 2 be two distinct finite local rings of length two with residue field of characteristic p . Let mathbb G ( mathcal O 2) and mathbb G ( mathcal O ' 2) , be the group of points of any reductive group scheme mathbb G over mathbb Z such that p is very good for mathbb G times mathbb F q . We prove that there exists an isomorphism of group algebra K mathbb G ( mathcal O 2) cong K mathbb G ( mathcal O ' 2) , where K is a sufficiently large field of characteristic different from p .
[ 11, 6 ]
physics.app-ph physics.flu-dyn
Joule Heating Induced Particle Manipulation on a Microfluidic Chip. We develop an electrokinetic technique that continuously manipulates colloidal particles to concentrate into patterned particulate groups in an energy efficient way, by exclusive harnessing of the intrinsic Joule heating effects. Our technique exploits the alternating current electrothermal flow phenomenon which is generated due to the interaction between non uniform electric and thermal fields. Highly non uniform electric field generates sharp temperature gradients by generating spatially varying Joule heat that varies along radial direction from a concentrated point hotspot. Sharp temperature gradients induce local variation in electric properties which, in turn, generate strong electrothermal vortex. The imposed fluid flow brings the colloidal particles at the centre of the hotspot and enables particle aggregation. Further, manoeuvering structures of the Joule heating spots, different patterns of particle clustering may be formed in a low power budget, thus, opening up a new realm of on chip particle manipulation process without necessitating highly focused laser beam which is much complicated and demands higher power budget. This technique can find its use in Lab on a chip devices to manipulate particle groups, including biological cells.
[ 13 ]
math.DS
Preservation of shadowing in discrete dynamical systems. We look at the preservation of various notions of shadowing in discrete dynamical systems under inverse limits, products, factor maps and the induced maps for symmetric products and hyperspaces. The shadowing properties we consider are the following shadowing, h shadowing, eventual shadowing, orbital shadowing, strong orbital shadowing, the first and second weak shadowing properties, limit shadowing, s limit shadowing, orbital limit shadowing and inverse shadowing.
[ 0 ]
cs.CV
Multi Frame Content Integration with a Spatio Temporal Attention Mechanism for Person Video Motion Transfer. Existing person video generation methods either lack the flexibility in controlling both the appearance and motion, or fail to preserve detailed appearance and temporal consistency. In this paper, we tackle the problem of motion transfer for generating person videos, which provides controls on both the appearance and the motion. Specifically, we transfer the motion of one person in a target video to another person in a source video, while preserving the appearance of the source person. Besides only relying on one source frame as the existing state of the art methods, our proposed method integrates information from multiple source frames based on a spatio temporal attention mechanism to preserve rich appearance details. In addition to a spatial discriminator employed for encouraging the frame level fidelity, a multi range temporal discriminator is adopted to enforce the generated video to resemble temporal dynamics of a real video in various time ranges. A challenging real world dataset, which contains about 500 dancing video clips with complex and unpredictable motions, is collected for the training and testing. Extensive experiments show that the proposed method can produce more photo realistic and temporally consistent person videos than previous methods. As our method decomposes the syntheses of the foreground and background into two branches, a flexible background substitution application can also be achieved.
[ 78, 15 ]
cs.IT math.AC math.IT
Multilinear Algebra for Minimum Storage Regenerating Codes. An (n, k, d, alpha) MSR (minimum storage regeneration) code is a set of n nodes used to store a file. For a file of total size k alpha , each node stores alpha symbols, any k nodes recover the file, and any d nodes can repair any other node via each sending out alpha (d k 1) symbols. In this work, we explore various ways to re express the infamous product matrix construction using skew symmetric matrices, polynomials, symmetric algebras, and exterior algebras. We then introduce a multilinear algebra foundation to produce bigl(n, k, frac (k 1)t t 1 , binom k 1 t 1 bigr) MSR codes for general t geq2 . At the t 2 end, they include the product matrix construction as a special case. At the t k end, we recover determinant codes of mode m k further restriction to n k 1 makes it identical to the layered code at the MSR point. Our codes' sub packetization level alpha is independent of n and small. It is less than L 2.8(d k 1) , where L is Alrabiah Guruswami's lower bound on alpha . Furthermore, it is less than other MSR codes' alpha for a subset of practical parameters. We offer hints on how our code repairs multiple failures at once.
[ 42 ]
math.DG
Cyclic conformally flat hypersurfaces revisited. In this article we classify the conformally flat Euclidean hypersurfaces of dimension three with three distinct principal curvatures of mathbb R 4 , mathbb S 3 times mathbb R and mathbb H 3 times mathbb R with the property that the tangent component of the vector field partial partial t is a principal direction at any point. Here partial partial t stands for either a constant unit vector field in mathbb R 4 or the unit vector field tangent to the factor mathbb R in the product spaces mathbb S 3 times mathbb R and mathbb H 3 times mathbb R , respectively. Then we use this result to give a simple proof of an alternative classification of the cyclic conformally flat hypersurfaces of mathbb R 4 , that is, the conformally flat hypersurfaces of mathbb R 4 with three distinct principal curvatures such that the curvature lines correspondent to one of its principal curvatures are extrinsic circles. We also characterize the cyclic conformally flat hypersurfaces of mathbb R 4 as those conformally flat hypersurfaces of dimension three with three distinct principal curvatures for which there exists a conformal Killing vector field of mathbb R 4 whose tangent component is an eigenvector field correspondent to one of its principal curvatures.
[ 49, 15 ]
math.OC cs.IT math.IT
Primal dual methods for large scale and distributed convex optimization and data analytics. The augmented Lagrangian method (ALM) is a classical optimization tool that solves a given "difficult" (constrained) problem via finding solutions of a sequence of "easier"(often unconstrained) sub problems with respect to the original (primal) variable, wherein constraints satisfaction is controlled via the so called dual variables. ALM is highly flexible with respect to how primal sub problems can be solved, giving rise to a plethora of different primal dual methods. The powerful ALM mechanism has recently proved to be very successful in various large scale and distributed applications. In addition, several significant advances have appeared, primarily on precise complexity results with respect to computational and communication costs in the presence of inexact updates and design and analysis of novel optimal methods for distributed consensus optimization. We provide a tutorial style introduction to ALM and its variants for solving convex optimization problems in large scale and distributed settings. We describe control theoretic tools for the algorithms' analysis and design, survey recent results, and provide novel insights in the context of two emerging applications federated learning and distributed energy trading.
[ 16, 82, 64 ]
q-bio.QM physics.soc-ph q-bio.MN
Learning Heat Diffusion for Network Alignment. Networks are abundant in the life sciences. Outstanding challenges include how to characterize similarities between networks, and in extension how to integrate information across networks. Yet, network alignment remains a core algorithmic problem. Here, we present a novel learning algorithm called evolutionary heat diffusion based network alignment (EDNA) to address this challenge. EDNA uses the diffusion signal as a proxy for computing node similarities between networks. Comparing EDNA with state of the art algorithms on a popular protein protein interaction network dataset, using four different evaluation metrics, we achieve (i) the most accurate alignments, (ii) increased robustness against noise, and (iii) superior scaling capacity. The EDNA algorithm is versatile in that other available network alignments embeddings can be used as an initial baseline alignment, and then EDNA works as a wrapper around them by running the evolutionary diffusion on top of them. In conclusion, EDNA outperforms state of the art methods for network alignment, thus setting the stage for large scale comparison and integration of networks.
[ 22 ]
cs.SE
OpenReq Issue Link Map A Tool to Visualize Issue Links in Jira. Managing software projects gets more and more complicated with an increasing project and product size. To cope with this complexity, many organizations use issue tracking systems, where tasks, bugs, and requirements are stored as issues. Unfortunately, managing software projects might remain chaotic even when using issue trackers. Particularly for long lasting projects with a large number of issues and links between them, it is often hard to maintain an overview of the dependencies, especially when dozens of new issues get reported every day. We present a Jira plug in that supports developers, project managers, and product owners in managing and overviewing issues and their dependencies. Our tool visualizes the issue links, helps to find missing or unknown links between issues, and detects inconsistencies.
[ 0 ]
cs.CV
Transformation Consistency Regularization A Semi Supervised Paradigm for Image to Image Translation. Scarcity of labeled data has motivated the development of semi supervised learning methods, which learn from large portions of unlabeled data alongside a few labeled samples. Consistency Regularization between model's predictions under different input perturbations, particularly has shown to provide state of the art results in a semi supervised framework. However, most of these method have been limited to classification and segmentation applications. We propose Transformation Consistency Regularization, which delves into a more challenging setting of image to image translation, which remains unexplored by semi supervised algorithms. The method introduces a diverse set of geometric transformations and enforces the model's predictions for unlabeled data to be invariant to those transformations. We evaluate the efficacy of our algorithm on three different applications image colorization, denoising and super resolution. Our method is significantly data efficient, requiring only around 10 20 of labeled samples to achieve similar image reconstructions to its fully supervised counterpart. Furthermore, we show the effectiveness of our method in video processing applications, where knowledge from a few frames can be leveraged to enhance the quality of the rest of the movie.
[ 9 ]
math.AP
Stability of the multi solitons of the modified Korteweg de Vries equation. We establish the nonlinear stability of N soliton solutions of the modified Korteweg de Vries (mKdV) equation. The N soliton solutions are global solutions of mKdV behaving at (positive and negative) time infinity as sums of 1 solitons with speeds 0 c 1 cdots c N .The proof relies on the variational characterization of N solitons. We show that the N solitons realize the local minimum of the (N 1) th mKdV conserved quantity subject to fixed constraints on the N first conserved quantities.To this aim, we construct a functional for which N solitons are critical points, we prove that the spectral properties of the linearization of this functional around a N soliton are preserved on the extended timeline, and we analyze the spectrum at infinity of linearized operators around 1 solitons. The main new ingredients in our analysis are a new operator identity based on a generalized Sylvester law of inertia and recursion operators for the mKdV equation.
[ 25, 34 ]
math.PR math.CV
Maximizing the p th moment of exit time of planar Brownian motion from a given domain. In this paper we address the question of finding the point which maximizes the p th moment of the exit time of planar Brownian motion from a given domain. We present a geometrical method of excluding parts of the domain from consideration which makes use of a coupling argument and the conformal invariance of Brownian motion. In many cases the maximizing point can be localized to a relatively small region. Several illustrative examples are presented.
[ 25 ]
math.PR
A Recursive Algorithm for Computing Inferences in Imprecise Markov Chains. We present an algorithm that can efficiently compute a broad class of inferences for discrete time imprecise Markov chains, a generalised type of Markov chains that allows one to take into account partially specified probabilities and other types of model uncertainty. The class of inferences that we consider contains, as special cases, tight lower and upper bounds on expected hitting times, on hitting probabilities and on expectations of functions that are a sum or product of simpler ones. Our algorithm exploits the specific structure that is inherent in all these inferences they admit a general recursive decomposition. This allows us to achieve a computational complexity that scales linearly in the number of time points on which the inference depends, instead of the exponential scaling that is typical for a naive approach.
[ 0 ]
cs.CV
DSNet An Efficient CNN for Road Scene Segmentation. Road scene understanding is a critical component in an autonomous driving system. Although the deep learning based road scene segmentation can achieve very high accuracy, its complexity is also very high for developing real time applications. It is challenging to design a neural net with high accuracy and low computational complexity. To address this issue, we investigate the advantages and disadvantages of several popular CNN architectures in terms of speed, storage and segmentation accuracy. We start from the Fully Convolutional Network (FCN) with VGG, and then we study ResNet and DenseNet. Through detailed experiments, we pick up the favorable components from the existing architectures and at the end, we construct a light weight network architecture based on the DenseNet. Our proposed network, called DSNet, demonstrates a real time testing (inferencing) ability (on the popular GPU platform) and it maintains an accuracy comparable with most previous systems. We test our system on several datasets including the challenging Cityscapes dataset (resolution of 1024x512) with an mIoU of about 69.1 and runtime of 0.0147 second per image on a single GTX 1080Ti. We also design a more accurate model but at the price of a slower speed, which has an mIoU of about 72.6 on the CamVid dataset.
[ 23 ]
math.FA
Fractional Biorthogonal wavelets in L 2( mathbb R) . The fractional Fourier transform (FrFT), which is a generalization of the Fourier transform, has become the focus of many research papers in recent years because of its applications in electrical engineering and optics. In this paper, we introduce the notion of fractional biorthogonal wavelets on mathbb R and obtain the necessary and sufficient conditions for the translates of a single function to form the fractional Riesz bases for their closed linear span. We also provide a complete characterization for the fractional biorthogonality of the translates of fractional scaling functions of two fractional MRAs and the associated fractional biorthogonal wavelet families. Moreover, under mild assumptions on the fractional scaling functions and the corresponding fractional wavelets, we show that the fractional wavelets can generate Reisz bases for L 2( mathbb R). .
[ 15 ]
cs.IT math.IT
Matrix Completion with Prior Subspace Information via Maximizing Correlation. This paper studies the problem of completing a low rank matrix from a few of its random entries with the aid of prior information. We suggest a strategy to incorporate prior information into the standard matrix completion procedure by maximizing the correlation between the original signal and the prior information. We also establish performance guarantees for the proposed method, which show that with suitable prior information, the proposed procedure can reduce the sample complexity of the standard matrix completion by a logarithmic factor. To illustrate the theory, we further analyze an important practical application where the prior subspace information is available. Both synthetic and real world experiments are provided to verify the validity of the theory.
[ 0 ]
cs.CV
Robust copy move forgery detection by false alarms control. Detecting reliably copy move forgeries is difficult because images do contain similar objects. The question is how to discard natural image self similarities while still detecting copy moved parts as being "unnaturally similar"? Copy move may have been performed after a rotation, a change of scale and followed by JPEG compression or the addition of noise. For this reason, we base our method on SIFT, which provides sparse keypoints with scale, rotation and illumination invariant descriptors. To discriminate natural descriptor matches from artificial ones, we introduce an a contrario method which gives theoretical guarantees on the number of false alarms. We validate our method on several databases. Being fully unsupervised it can be integrated into any generic automated image tampering detection pipeline.
[ 0 ]
cs.CV
Dunhuang Grottoes Painting Dataset and Benchmark. This document introduces the background and the usage of the Dunhuang Grottoes Dataset and the benchmark. The documentation first starts with the background of the Dunhuang Grotto, which is widely recognised as an priceless heritage. Given that digital method is the modern trend for heritage protection and restoration. Follow the trend, we release the first public dataset for Dunhuang Grotto Painting restoration. The rest of the documentation details the painting data generation. To enable a data driven fashion, this dataset provided a large number of training and testing example which is sufficient for a deep learning approach. The detailed usage of the dataset as well as the benchmark is described.
[ 41 ]
cs.RO
Robot Learning of Shifting Objects for Grasping in Cluttered Environments. Robotic grasping in cluttered environments is often infeasible due to obstacles preventing possible grasps. Then, pre grasping manipulation like shifting or pushing an object becomes necessary. We developed an algorithm that can learn, in addition to grasping, to shift objects in such a way that their grasp probability increases. Our research contribution is threefold First, we present an algorithm for learning the optimal pose of manipulation primitives like clamping or shifting. Second, we learn non prehensible actions that explicitly increase the grasping probability. Making one skill (shifting) directly dependent on another (grasping) removes the need of sparse rewards, leading to more data efficient learning. Third, we apply a real world solution to the industrial task of bin picking, resulting in the ability to empty bins completely. The system is trained in a self supervised manner with around 25000 grasp and 2500 shift actions. Our robot is able to grasp and file objects with 274 picks per hour. Furthermore, we demonstrate the system's ability to generalize to novel objects.
[ 1, 2, 3 ]
cs.CL cs.AI cs.LG
A Neural Topical Expansion Framework for Unstructured Persona oriented Dialogue Generation. Unstructured Persona oriented Dialogue Systems (UPDS) has been demonstrated effective in generating persona consistent responses by utilizing predefined natural language user persona descriptions (e.g., "I am a vegan"). However, the predefined user persona descriptions are usually short and limited to only a few descriptive words, which makes it hard to correlate them with the dialogues. As a result, existing methods either fail to use the persona description or use them improperly when generating persona consistent responses. To address this, we propose a neural topical expansion framework, namely Persona Exploration and Exploitation (PEE), which is able to extend the predefined user persona description with semantically correlated content before utilizing them to generate dialogue responses. PEE consists of two main modules persona exploration and persona exploitation. The former learns to extend the predefined user persona description by mining and correlating with existing dialogue corpus using a variational auto encoder (VAE) based topic model. The latter learns to generate persona consistent responses by utilizing the predefined and extended user persona description. In order to make persona exploitation learn to utilize user persona description more properly, we also introduce two persona oriented loss functions Persona oriented Matching (P Match) loss and Persona oriented Bag of Words (P BoWs) loss which respectively supervise persona selection in encoder and decoder. Experimental results show that our approach outperforms state of the art baselines, in terms of both automatic and human evaluations.
[ 38 ]
math.NT
Hyperbolicity of the partition Jensen polynomials. Given an arithmetic function a mathbb N rightarrow mathbb R , one can associate a naturally defined, doubly infinite family of Jensen polynomials. Recent work of Griffin, Ono, Rolen, and Zagier shows that for certain families of functions a mathbb N rightarrow mathbb R , the associated Jensen polynomials are eventually hyperbolic (i.e., eventually all of their roots are real). This work proves Chen, Jia, and Wang's conjecture that the partition Jensen polynomials are eventually hyperbolic as a special case. Here, we make this result explicit. Let N(d) be the minimal number such that for all n geq N(d) , the partition Jensen polynomial of degree d and shift n is hyperbolic. We prove that N(3) 94 , N(4) 206 , and N(5) 381 , and in general, that N(d) leq (3d) 24d (50d) 3d 2 .
[ 9 ]
math.AP
Operator error estimates for homogenization of the nonstationary Schr " o dinger type equations sharpness of the results. In L 2 ( mathbb R d mathbb C n) , we consider a selfadjoint matrix strongly elliptic second order differential operator mathcal A varepsilon with periodic coefficients depending on mathbf x varepsilon . We find approximations of the exponential e i tau mathcal A varepsilon , tau in mathbb R , for small varepsilon in the ( H s to L 2 ) operator norm with suitable s . The sharpness of the error estimates with respect to tau is discussed. The results are applied to study the behavior of the solution mathbf u varepsilon of the Cauchy problem for the Schr " o dinger type equation i partial tau mathbf u varepsilon mathcal A varepsilon mathbf u varepsilon mathbf F .
[ 68, 21 ]
math.CO cs.DM
Explicit expanders of every degree and size. An (n,d, lambda) graph is a d regular graph on n vertices in which the absolute value of any nontrivial eigenvalue is at most lambda . For any constant d geq 3 , epsilon 0 and all sufficiently large n we show that there is a deterministic poly(n) time algorithm that outputs an (n,d, lambda) graph (on exactly n vertices) with lambda leq 2 sqrt d 1 epsilon . For any d p 2 with p equiv 1 bmod 4 prime and all sufficiently large n , we describe a strongly explicit construction of an (n,d, lambda) graph (on exactly n vertices) with lambda leq sqrt 2(d 1) sqrt d 2 o(1) ( (1 sqrt 2) sqrt d 1 o(1)) , with the o(1) term tending to 0 as n tends to infinity. For every epsilon 0 , d d 0( epsilon) and n n 0(d, epsilon) we present a strongly explicit construction of an (m,d, lambda) graph with lambda (2 epsilon) sqrt d and m n o(n) . All constructions are obtained by starting with known ones of Ramanujan or nearly Ramanujan graphs, modifying or packing them in an appropriate way. The spectral analysis relies on the delocalization of eigenvectors of regular graphs in cycle free neighborhoods.
[ 65, 19 ]
cs.NI cs.CR
Security, Privacy and Ethical Concerns of IoT Implementations in Hospitality Domain. The Internet of Things (IoT) has been on the rise in the last decade as it finds applications in various domains. Hospitality is one of the pioneer sectors that has adopted this technology to create novel services such as smart hotel rooms, personalized services etc. Hotels, restaurants, theme parks, and cruise ships are some specific application areas to improve customer satisfaction by creating an intense interactive environment and data collection with the use of appropriate sensors and actuators. However, applying IoT solutions in the hospitality environment has some unique challenges such as easy physical access to devices. In addition, due to the very nature of these domains, the customers are at the epicenter of these IoT technologies that result in a massive amount of data collection from them. Such data and its management along with business purposes also raises new concerns regarding privacy and ethical considerations. Therefore, this paper surveys and analyzes security, privacy and ethical issues regarding the utilization of IoT devices by focusing on the hospitality industry specifically. We explore some exemplary uses, cases, potential problems and solutions in order to contribute to better understanding and guiding the business operators in this sector.
[ 40, 3 ]
cs.CL cs.IR
BioNerFlair biomedical named entity recognition using flair embedding and sequence tagger. Motivation The proliferation of Biomedical research articles has made the task of information retrieval more important than ever. Scientists and Researchers are having difficulty in finding articles that contain information relevant to them. Proper extraction of biomedical entities like Disease, Drug chem, Species, Gene protein, can considerably improve the filtering of articles resulting in better extraction of relevant information. Performance on BioNer benchmarks has progressively improved because of progression in transformers based models like BERT, XLNet, OpenAI, GPT2, etc. These models give excellent results however, they are computationally expensive and we can achieve better scores for domain specific tasks using other contextual string based models and LSTM CRF based sequence tagger. Results We introduce BioNerFlair, a method to train models for biomedical named entity recognition using Flair plus GloVe embeddings and Bidirectional LSTM CRF based sequence tagger. With almost the same generic architecture widely used for named entity recognition, BioNerFlair outperforms previous state of the art models. I performed experiments on 8 benchmarks datasets for biomedical named entity recognition. Compared to current state of the art models, BioNerFlair achieves the best F1 score of 90.17 beyond 84.72 on the BioCreative II gene mention (BC2GM) corpus, best F1 score of 94.03 beyond 92.36 on the BioCreative IV chemical and drug (BC4CHEMD) corpus, best F1 score of 88.73 beyond 78.58 on the JNLPBA corpus, best F1 score of 91.1 beyond 89.71 on the NCBI disease corpus, best F1 score of 85.48 beyond 78.98 on the Species 800 corpus, while near best results was observed on BC5CDR chem, BC3CDR disease, and LINNAEUS corpus.
[ 0, 1, 2 ]
cs.LG cs.AI cs.CV
Identifying and interpreting tuning dimensions in deep networks. In neuroscience, a tuning dimension is a stimulus attribute that accounts for much of the activation variance of a group of neurons. These are commonly used to decipher the responses of such groups. While researchers have attempted to manually identify an analogue to these tuning dimensions in deep neural networks, we are unaware of an automatic way to discover them. This work contributes an unsupervised framework for identifying and interpreting "tuning dimensions" in deep networks. Our method correctly identifies the tuning dimensions of a synthetic Gabor filter bank and tuning dimensions of the first two layers of InceptionV1 trained on ImageNet.
[ 79 ]
physics.hist-ph
Are there really many worlds in the "Many worlds interpretation" of Quantum Mechanics?. Since the 1970s, the Everett Wheeler many worlds interpretation (MWI) of Quantum Mechanics (1955) has been much in the news. One wonders about the worlds in question, their branches, their splittings, their number. It is most often ignored that this language is not that of Everett, whom Wheeler very quickly stopped supporting. Moreover, for some interpreters, the real meaning of Everett ideas is not the coexistence of many worlds, but the existence of a single quantum one. In this context, what should we think of attempts to verify Everett thesis? What about the connexion between MWI and the cosmological multiverse? How to understand the links between the quantum world and the classical one? This article tries to answer some of these questions.
[ 49 ]
math.OC
Local null controllability of a nonlocal semilinear heat equation. This paper deals with the problem of internal null controllability of a heat equation posed on a bounded domain with Dirichlet boundary conditions and perturbed by a semilinear nonlocal term. We prove the small time local null controllability of the equation. The proof relies on two main arguments. First, we establish the small time local null controllability of a 2 times 2 reaction diffusion system, where the second equation is governed by the parabolic operator tau partial t sigma Delta , tau, sigma 0 . More precisely, this controllability result is obtained uniformly with respect to the parameters ( tau, sigma) in (0,1) times (1, infty) . Secondly, we observe that the semilinear nonlocal heat equation is actually the asymptotic derivation of the reaction diffusion system in the limit ( tau, sigma) rightarrow (0, infty) . Finally, we illustrate these results by numerical simulations.
[ 40, 1, 3 ]
cs.CL cs.IR cs.LG
KINNEWS and KIRNEWS Benchmarking Cross Lingual Text Classification for Kinyarwanda and Kirundi. Recent progress in text classification has been focused on high resource languages such as English and Chinese. For low resource languages, amongst them most African languages, the lack of well annotated data and effective preprocessing, is hindering the progress and the transfer of successful methods. In this paper, we introduce two news datasets (KINNEWS and KIRNEWS) for multi class classification of news articles in Kinyarwanda and Kirundi, two low resource African languages. The two languages are mutually intelligible, but while Kinyarwanda has been studied in Natural Language Processing (NLP) to some extent, this work constitutes the first study on Kirundi. Along with the datasets, we provide statistics, guidelines for preprocessing, and monolingual and cross lingual baseline models. Our experiments show that training embeddings on the relatively higher resourced Kinyarwanda yields successful cross lingual transfer to Kirundi. In addition, the design of the created datasets allows for a wider use in NLP beyond text classification in future studies, such as representation learning, cross lingual learning with more distant languages, or as base for new annotations for tasks such as parsing, POS tagging, and NER. The datasets, stopwords, and pre trained embeddings are publicly available at https github.com Andrews2017 KINNEWS and KIRNEWS Corpus .
[ 46, 38 ]
math.NT math.AG
The average size of the 2 Selmer group of a family of non hyperelliptic curves of genus 3. We show that the average size of the 2 Selmer group of the family of Jacobians of non hyperelliptic genus 3 curves with a marked rational hyperflex point, when ordered by a natural height, is bounded above by 3 . We achieve this by interpreting 2 Selmer elements as integral orbits of a representation associated with a stable mathbb Z 2 mathbb Z grading on the Lie algebra of type E 6 and using Bhargava's orbit counting techniques. We use this result to show that the marked point is the only rational point for a positive proportion of curves in this family. The main novelties are the construction of integral representatives using certain properties of the compactified Jacobian of the simple curve singularity of type E 6 , and a representation theoretic interpretation of a Mumford theta group naturally associated to our family of curves.
[ 65 ]
cs.NI
On Extending the Wireless Communications Range of Weather Stations using LoRaWAN. Consumer grade weather stations typically involve transmitting sensor information to a digital console that provides readouts of the data being collected. The wireless range of such system is confined to about 100 m, making their use limited to urban environments. We present a device that decodes the data being sent by the weather station and forwards them using the emerging LoRaWAN technology. We designed the device keeping in mind the peculiar conditions of Developing Countries, in particular the low cost and low power requirements. This allows interesting applications in the realm of disaster prevention and mitigation using a network of numerous weather stations.
[ 26 ]
math.RT
Representation theory of a semisimple extension of the Takiff superalgebra. We study a semisimple extension of a Takiff superalgebra which turns out to have a remarkably rich representation theory. We determine the blocks in both the finite dimensional and BGG module categories and also classify the Borel subalgebras. We further compute all extension groups between two finite dimensional simple objects and prove that all non principal blocks in the finite dimensional module category are Koszul.
[ 27 ]
math.CT
Algebraically cofibrant and fibrant objects revisited. We extend all known results about transferred model structures on algebraically cofibrant and fibrant objects by working with weak model categories. We show that for an accessible weak model category there are always Quillen equivalent transferred weak model structures on both the categories of algebraically cofibrant and algebraically fibrant objects. Under additional assumptions, these transferred weak model structures are shown to be left, right or Quillen model structures. By combining both constructions, we show that each combinatorial weak model category is connected, via a chain of Quillen equivalences, to a combinatorial Quillen model category in which all objects are fibrant.
[ 0 ]
cs.CV
Transferrable Prototypical Networks for Unsupervised Domain Adaptation. In this paper, we introduce a new idea for unsupervised domain adaptation via a remold of Prototypical Networks, which learn an embedding space and perform classification via a remold of the distances to the prototype of each class. Specifically, we present Transferrable Prototypical Networks (TPN) for adaptation such that the prototypes for each class in source and target domains are close in the embedding space and the score distributions predicted by prototypes separately on source and target data are similar. Technically, TPN initially matches each target example to the nearest prototype in the source domain and assigns an example a "pseudo" label. The prototype of each class could then be computed on source only, target only and source target data, respectively. The optimization of TPN is end to end trained by jointly minimizing the distance across the prototypes on three types of data and KL divergence of score distributions output by each pair of the prototypes. Extensive experiments are conducted on the transfers across MNIST, USPS and SVHN datasets, and superior results are reported when comparing to state of the art approaches. More remarkably, we obtain an accuracy of 80.4 of single model on VisDA 2017 dataset.
[ 35, 13 ]
math.DS math.RA
Ring of flows of one dimensional differential equations. In this article, the ring of flows of autonomous differential equations of order one on integral domains is constructed. First, we build the autonomous ring Opa( hurw R x ) and then its structure is studied. Next, we build the ring of formal exponential generating series of the ring Opa( hurw R x ) , where it is possible to find solutions of differential equations of order one when the vector field of the system can be decomposed in sum or product of functions.
[ 46 ]
math.AG
On the Inertia Conjecture and its generalizations. Studying two point branched Galois covers of the projective line we prove the Inertia Conjecture for the Alternating groups A p 1 , A p 3 , A p 4 for any odd prime p equiv 2 pmod 3 and for the group A p 5 when additionally 4 nmid (p 1) and p geq 17 . We obtain a generalization of a patching result by Raynaud which reduces these proofs to showing the realizations of the ' e tale Galois covers of the affine line with a fewer candidates for the inertia groups above infty . We also pose a general question motivated by the Inertia Conjecture and obtain some affirmative results. A special case of this question, which we call the Generalized Purely Wild Inertia Conjecture, is shown to be true for the groups for which the purely wild part of the Inertia Conjecture is already established. In particular, we show that if this generalized conjecture is true for the groups G 1 and G 2 which do not have a common quotient, then the conjecture is also true for the product G 1 times G 2 .
[ 20 ]
math.NA cs.NA
A novel greedy Gauss Seidel method for solving large linear least squares problem. We present a novel greedy Gauss Seidel method for solving large linear least squares problem. This method improves the greedy randomized coordinate descent (GRCD) method proposed recently by Bai and Wu Bai ZZ, and Wu WT. On greedy randomized coordinate descent methods for solving large linear least squares problems. Numer Linear Algebra Appl. 2019 26(4) 1 15 , which in turn improves the popular randomized Gauss Seidel method. Convergence analysis of the new method is provided. Numerical experiments show that, for the same accuracy, our method outperforms the GRCD method in term of the computing time.
[ 0 ]
cs.CV
Domain Adaptation for Semantic Segmentation with Maximum Squares Loss. Deep neural networks for semantic segmentation always require a large number of samples with pixel level labels, which becomes the major difficulty in their real world applications. To reduce the labeling cost, unsupervised domain adaptation (UDA) approaches are proposed to transfer knowledge from labeled synthesized datasets to unlabeled real world datasets. Recently, some semi supervised learning methods have been applied to UDA and achieved state of the art performance. One of the most popular approaches in semi supervised learning is the entropy minimization method. However, when applying the entropy minimization to UDA for semantic segmentation, the gradient of the entropy is biased towards samples that are easy to transfer. To balance the gradient of well classified target samples, we propose the maximum squares loss. Our maximum squares loss prevents the training process being dominated by easy to transfer samples in the target domain. Besides, we introduce the image wise weighting ratio to alleviate the class imbalance in the unlabeled target domain. Both synthetic to real and cross city adaptation experiments demonstrate the effectiveness of our proposed approach. The code is released at https github. com ZJULearning MaxSquareLoss.