## COVID 19 Diagnosis from Cough Acoustics using ConvNets and Data Augmentation

With the periodic rise and fall of COVID-19 and countries being inflicted byits waves, an efficient, economic, and effortless diagnosis procedure for the virus has been the utmost need of the hour . Thesedifferences in the coughing sounds are minute and indiscernible to the humanear, however, these can be captured using machine learning-based statistical models .…

## Complexity of optimizing over the integers

In the first part of this paper, we present a unified framework for analyzing the algorithmic complexity of any optimization problem, whether it becontinuous or discrete in nature . This helps to formalize notions like “input”,”size” and “complexity” in the context of general mathematical optimization .…

## GraPE fast and scalable Graph Processing and Embedding

Graph Representation Learning methods have enabled a wide range of learning problems to be addressed for data that can be represented in graph form . Several real world problems in economy, biology, medicine and other fields raised relevant scaling problems with existing methods and their implementation .…

## Mutual cooperation and tolerance to defection in the context of socialization the theoretical model and experimental evidence

Previous findings reveal that socialization effectively promotes cooperation in the well-known Prisoner’s dilemma (PD) game . However, theoretical concepts fail to describe high levels of cooperation (probability higher than 50%) that were observed empirically . In this paper, we derive asymmetrical quantal response equilibrium (QRE) in PD in Markov strategies andtest it against experimental data .…

## Fine grained style control in Transformer based Text to speech Synthesis

In this paper, we present a novel architecture to realize fine-grained stylecontrol on the transformer-based text-to-speech synthesis (TransformerTTS) We model the speaking style by extracting a time sequence of local style tokens (LST) from the reference speech . We prevent style embedding from encoding linguistic content by randomly truncating LST during training and using wav2vec 2.0features .…

## Seamless Copy Move Manipulation in Digital Images

The method exploits the localized nature of discrete wavelettransform (DWT) to get hold of the region of the host image to be manipulated . The proposed method shows good resistance againstdetection by two frequency domain forgery detection methods from theliterature .…

## An Annihilating Filter Based DOA Estimation for Uniform Linear Array

In this paper, we propose a new method to design an annihilating filter (AF) for direction-of-arrival (DOA) estimation of multiple snapshots within anuniform linear array . To evaluate the proposed method, we firstly design a DOAestimation using multiple signal classification (MUSIC) algorithm, referred toas the MUSIC baseline .…

## A closest point method library for PDEs on surfaces with parallel domain decomposition solvers and preconditioners

DD-CPM software library provides a set of tools for the discretizationand solution of problems arising from the closest point method (CPM) forpartial differential equations on surfaces . The software is detailed herein, and a number of problems and benchmarks are demonstrated .…

## Delay Sensitive and Power Efficient Quality Control of Dynamic Video Streaming using Adaptive Super Resolution

A novel dynamic video streaming algorithm that compresses video chunks at the transmitter and separately enhancesthe quality at the receiver using SR . Simulation results show that the proposed scheme pursues the quality-of-services (QoS) of the video streaming better than the adaptive quality control without the cooperation of the transmitters and the receiver .…

## Codabench Flexible Easy to Use and Reproducible Benchmarking for Everyone

Codabench is an open-sourced, community-driven platform for benchmarking algorithms or software agents versus datasets or tasks . It supports code submission and datasubmission for testing on dedicated compute workers, which can be supplied by benchmark organizers . This makes the system scalable, at low cost for the platform providers .…

## Diagonalizing Against Polynomial Time Bounded Turing Machines Via Nondeterministic Turing Machine

The diagonalization technique was invented by Cantor to show that there are more real numbers than algebraic numbers . In this work, we enumerate all polynomial-time deterministic Turing machines and diagonalize over all of them by an universal nondeterministic Turing machine .…

## A large scale lexical and semantic analysis of Spanish language variations in Twitter

Dialectometry is a discipline devoted to studying the variations of alanguage around a geographical region . For example, Spanish is one of themost spoken languages across the world, but not necessarily Spanish is written in the same way in different countries .…

## Mesh Draping Parametrization Free Neural Mesh Transfer

Mesh Draping is a neural method for transferring existing meshstructure from one shape to another . The method drapes the source mesh over the target geometry and at the same time seeks to preserve the carefully designedcharacteristics of the source .…

## Learning to Coordinate in Multi Agent Systems A Coordinated Actor Critic Algorithm and Finite Time Guarantees

Multi-agent reinforcement learning (MARL) has attracted much researchattention recently . Many theoretical and algorithmic aspects of MARL have not been well-understood . We propose and analyze a class of coordinated actor-critic algorithms (CAC) in which individuallyparametrized policies have a shared part .…

## MetricGAN U Unsupervised speech enhancement dereverberation based only on noisy reverberated speech

In MetricGAN-U, only noisy speech is required to train the model by optimizing non-intrusivespeech quality metrics . The experimental results verified that the model outperforms baselines in both objective and subjective metrics . In this paper, wepropose Metricgan-U to further release the constraint from conventional unsupervised learning .…

## Fast Block Linear System Solver Using Q Learning Schduling for Unified Dynamic Power System Simulations

A fast block direct solver for the unified dynamic simulations of power systems . This solver uses a novel Q-learning based method for taskscheduling . The simulation on some large power systems shows that our solveris 2-6 times faster than KLU .…

## Dimensionality Reduction for k Distance Applied to Persistent Homology

Given a set P of n points and a constant k, we are interested in computing the persistent homology of the Cech filtration of P for the k-distance . We show that anylinear transformation that preserves pairwise distances up to a (1 +/- e)multiplicative factor, must preserve the persistent .…

## Optimal rate of convergence for approximations of SPDEs with non regular drift

A fully discrete finite difference scheme for stochastic reaction-diffusionequations driven by a $1+1$-dimensional white noise is studied . The optimalstrong rate of convergence is proved without posing any regularity assumption .…

## Predictive design of impact absorbers for mitigating resonances of flexible structures using a semi analytical approach

Analytical conditions are available for the optimum design of impactabsorbers for the case where the host structure is well described as rigidbody . The analysis relies on the assumption that the impacts cause immediate dissipation in the contact region, which is modeled in terms of a known coefficient of restitution .…

## MELONS generating melody with long term structure using transformers and structure graph

The creation of long melody sequences requires effective expression of musical structure . MELONS adopts a multi-step generation method with transformer-based networks by factoring melody generation into two sub-problems: structuregeneration and structure conditional melody generation . Experimental resultsshow that MELON can produce structured melodies with high quality and richcontents.…

## Synergy Resource Sensitive DNN Scheduling in Multi Tenant Clusters

Training Deep Neural Networks (DNNs) is a widely popular workload in bothenterprises and cloud data centers . Existing schedulers for DNN training allocate resources proportional to the number of GPUs requested by the job . In this work,we propose Synergy, a resource-sensitive scheduler for shared GPU clusters .…

## Inexact IETI DP for conforming isogeometric multi patch discretizations

In this paper, we investigate Dual-Primal Isogeometric Tearing andInterconnecting (IETI-DP) methods for conforming Galerkin discretizations on multi-patch computational domains with inexact subdomain solvers . We replace sparse LU factorizations by fast diagonalization based preconditioners to get a faster IETi-DP method while maintaining the same condition number bound .…

## Directionality Reinforcement Learning to Operate Multi Agent System without Communication

This paper establishes directionality reinforcement learning (DRL) techniqueto propose the complete decentralized multi-agent reinforcement learning method . Concretely, DRL adds the direction “agents have to learn toreach the farthest goal among reachable ones” to learning agents to operate theagents cooperatively .…

## GraPE fast and scalable Graph Processing and Embedding

Graph Representation Learning methods have enabled a wide range of learning problems to be addressed for data that can be represented in graph form . Several real world problems in economy, biology, medicine and other fields raised relevant scaling problems with existing methods and their implementation .…

## An enumeration of 1 perfect ternary codes

We study codes with parameters of the ternary Hamming $(n,3^{n-m), 3)$ code . The rank of the code is defined as thedimension of its affine span . We count their number and prove that all such codes can be obtained from each other by a sequence of two-coordinate switchings .…

## Impact of delay classes on the data structure in IOTA

In distributed ledger technologies (DLTs) with a directed acyclic graph (DAG) data structure, a message-issuing node can decide where to append that message and how to grow the DAG . The selection of the parent messages to which a node append depends on which messages it considers as tips .…

## Improved Analysis of EDCS via Gallai Edmonds Decomposition

An EDCS is a sparse subgraphsatisfying simple edge-degree constraints that is guaranteed to include an(almost) $\frac{2}{3)$-approximate matching of the base graph . Since itsintroduction, the EDCS has been successfully applied to numerous models ofcomputation . Motivated by this success, we revisit EDCS and present an improvedbound for its key property in general graphs .…

## Photonic Networks on Chip Employing Multilevel Signaling A Cross Layer Comparative Study

Photonic network-on-chip (PNoC) architectures employ photonic links withdense wavelength-division multiplexing (DWDM) to enable high throughput on-chiptransfers . Increasing the DWDM degree (i.e., using a largernumber of wavelengths) requires sophisticated and costly lasersources along with extra photonic hardware . This extra hardware can introduceundesired noise to the photonic link and increase the bit-error-rate (BER), power, and area consumption of PNoCs .…

## Inclusive Design Accessibility Settings for People with Cognitive Disabilities

The advancement of technology has progressed faster than any other field inthe world and with the development of these new technologies, it is important to make sure that these tools can be used by everyone, including people with disabilities . The purpose of this paper is to suggest amore affordable and readily available option for ALS assistive technology that can be implemented on a smartphone or tablet .…

## Complexity of optimizing over the integers

In the first part of this paper, we present a unified framework for analyzing the algorithmic complexity of any optimization problem, whether it becontinuous or discrete in nature . This helps to formalize notions like “input”,”size” and “complexity” in the context of general mathematical optimization .…

## A Modeling Framework for Efficient Reduced Order Simulations of Parametrized Lithium Ion Battery Cells

In this contribution we present a new modeling and simulation framework for Lithium-ion battery cells . We first derive a new continuum model for a rather general intercalation battery cell on the basis of non-equilibriumthermodynamics . The reduced basis method is a model order reduction technique on thebasis of an incremental hierarchical approximate proper orthogonaldecomposition approach and empirical operator interpolation .…

## Inclusive Design Accessibility Settings for People with Cognitive Disabilities

The advancement of technology has progressed faster than any other field inthe world and with the development of these new technologies, it is important to make sure that these tools can be used by everyone, including people with disabilities . The purpose of this paper is to suggest amore affordable and readily available option for ALS assistive technology that can be implemented on a smartphone or tablet .…

## Boundary integral equation methods for the solution of scattering and transmission 2D elastodynamic problems

We introduce and analyze various Regularized Combined Field IntegralEquations (CFIER) formulations of time-harmonic Navier equations in media withpiece-wise constant material properties . We also use the DtN approximations to deriveand analyze Optimized Schwarz (OS) methods for the solution of elastodynamicstransmission problems .…

## Couple Learning Mean Teacher method with pseudo labels improves semi supervised deep learning results

The recently proposed Mean Teacher has achieved state-of-the-art results in semi-supervised learning benchmarks . The Couple Learning method can extract more information inthe compound training data . The proposed pseudo-labels generated model (PLG) canincrease strongly-labeled data and weakly-labeled data to improve performance of the Mean Teacher method .…

## Deep Federated Learning for Autonomous Driving

Autonomous driving is an active research topic in both academia and industry . We propose a peer-to-peer Deep Federated Learning (DFL) approach totrain deep architectures in a fully decentralized manner and remove the need for central orchestration . We design a new Federated Autonomous Driving network(FADNet) that can improve the model stability, ensure convergence, and handleimbalanced data distribution problems while is being trained with federatedlearning methods .…

## Open Player Modeling Empowering Players through Data Transparency

Data is becoming an important central point for making design decisions formost software . As data-driven methods and systems start to populate these environments, a good question is: can we make models developed from this data transparent to users? In this paper, wesynthesize existing work from the Intelligent User Interface and LearningScience research communities, where they started to investigate the potential of making such data and models available to users .…

## Fast Block Linear System Solver Using Q Learning Schduling for Unified Dynamic Power System Simulations

A fast block direct solver for the unified dynamic simulations of power systems . This solver uses a novel Q-learning based method for taskscheduling . The simulation on some large power systems shows that our solveris 2-6 times faster than KLU .…

## Structural invariants in individuals language use the ego network of words

The cognitive constraints that humans exhibit in their social interactions have been extensively studied by anthropologists . We postulate thatsimilar regularities can be found in other cognitive processes such as language production . We analyse a dataset containing tweets of a heterogeneous group of Twitter users (regular users and professional writers) Leveraging amethodology similar to the one used to uncover the well-established socialcognitive constraints, we find that a concentric layered structure (which wecall ego network of words) very well captures how individuals organise the words they use .…

## Delay Sensitive and Power Efficient Quality Control of Dynamic Video Streaming using Adaptive Super Resolution

A novel dynamic video streaming algorithm that compresses video chunks at the transmitter and separately enhancesthe quality at the receiver using SR . Simulation results show that the proposed scheme pursues the quality-of-services (QoS) of the video streaming better than the adaptive quality control without the cooperation of the transmitters and the receiver .…

## Reverse Engineering Code Dependencies Converting Integer Based Variability to Propositional Logic

This paper introduces an approach to convert integer-basedvariability conditions to propositional logic . It works well on restricted variables (i.e.variables with a small range of allowed values) But unrestricted integer variables are handled less exact, but still retain useful variability information .…

## Secure Email A Usability Study

More than 60% of e-mail users are unaware of theexistence of end-to-end encryption technologies and never tried to use one . Users are overwhelmed with the management of publickeys and struggle with the setup of encryption technology in their mail software .…

## Complexity of optimizing over the integers

In the first part of this paper, we present a unified framework for analyzing the algorithmic complexity of any optimization problem, whether it becontinuous or discrete in nature . This helps to formalize notions like “input”,”size” and “complexity” in the context of general mathematical optimization .…

## A Time Encoding approach to training Spiking Neural Networks

Spiking Neural Networks (SNNs) have been gaining in popularity, it seems that the algorithms used to train them are not powerful enough to solve the same tasks as those tackled by classical Artificial Neural Networks . In this paper, we provide an extra tool to help us understand and train SNNs by using theory from the field of time encoding .…

## Finding Relevant Points for Nearest Neighbor Classification

In nearest-neighbor classification problems, a set of $d$-dimensional training points are used to infer unknown classifications of other points . A training point is relevant if its omission from the training set would change the outcome of some of these inferences .…

## A Categorical Semantics of Fuzzy Concepts in Conceptual Spaces

We propose log-concave functions as models of fuzzy concepts . We define a symmetric monoidal category modelling fuzzy concepts and fuzzyconceptual reasoning . This allows one to model fuzzy reasoning with noisy inputs, and provides a novel example of a Markov category .…

## Provably Efficient Reinforcement Learning in Decentralized General Sum Markov Games

This paper addresses the problem of learning an equilibrium efficiently ingeneral-sum Markov games through decentralized multi-agent reinforcementlearning . We propose an algorithm in which each agent independently runs optimistic V-learning (a variant of Q-learning) to explore the unknown environment, while using a stabilized onlinemirror descent (OMD) subroutine for policy updates .…

## MELONS generating melody with long term structure using transformers and structure graph

The creation of long melody sequences requires effective expression of musical structure . MELONS adopts a multi-step generation method with transformer-based networks by factoring melody generation into two sub-problems: structuregeneration and structure conditional melody generation . Experimental resultsshow that MELON can produce structured melodies with high quality and richcontents.…

## As Easy as ABC Adaptive Binning Coincidence Test for Uniformity Testing

We consider the problem of uniformity testing of Lipschitz continuousdistributions with bounded support . We propose a sequential test that adapts to the unknown distribution under the alternative hypothesis . The ABC test builds on a novel sequentialcoincidence test for discrete distributions, which is of independent interest .…

## Incremental Community Detection in Distributed Dynamic Graph

In this paper, we propose an incremental community detection algorithm for maintaining a dynamic graph over streaming data . The results demonstrate that our IDWCC algorithm performs up tothree times faster than the DWCC algorithm for a similar accuracy . Wevalidate the functionality and efficiency of our framework in processingstreaming data and performing large in-memory distributed dynamic graphanalytics .…

## LazySets jl Scalable Symbolic Numeric Set Computations

LazySets.jl is an easy-to-use, general-purpose and scalable library for computations that mix symbolics and numerics . It is the core library of JuliaReach, a cutting-edge software addressing the fundamental problem of reachability analysis . The library was originally designed for reachability and formal verification, but its scope goes beyond such topics .…