## Understanding and Achieving Efficient Robustness with Adversarial Contrastive Learning

Contrastive learning (CL) has recently emerged as an effective approach to learning representation in a range of downstream tasks . Central to this approach is the selection of positive (similar) and negative (dissimilar) setsto provide the model the opportunity to contrast’ between data and classrepresentation in the latent space .…

## The emergence of visual semantics through communication games

In this work, we consider a signalling gamesetting in which a sender’ agent must communicate the information about animage to a `receiver’ who must select the correct image from many distractors . We investigate the effect of the feature extractor’s weights and of the task being solved on the visual semantics learned by the models .…

## GP Context free Grammar Pre training for Text to SQL Parsers

A new method for Text-to-SQL parsing, Grammar Pre-training (GP) is proposed to decode deep relations between question and database . A random value is added behind a questionword which is recognized as a column, and the new sentence serves as the model input .…

## Spectral Leakage and Rethinking the Kernel Size in CNNs

Convolutional layers in CNNs implement linear filters which decompose the input into different frequency bands . We show that the small size of CNN kernels make them susceptible to spectral leakage . To address this issue, we propose theuse of larger kernel sizes along with the Hamming window function to alleviateleakage in CNN architectures .…

## Weakly Supervised Learning for Facial Behavior Analysis A Review

In recent years, there has been a shift in facial behavior analysis from the laboratory-controlled conditions to the challenging in-the-wild conditions due to superior performance of deep learning based approaches for many realworld applications . Labeling process of huge training data demands lot of human support with strong domainexpertise for facial expressions or action units, which is difficult to obtainin real-time environments .…

## Parametric Rectified Power Sigmoid Units Learning Nonlinear Neural Transfer Analytical Forms

The paper proposes representation functionals in a dual paradigm where learning jointly concerns both linear convolutional weights and parametricforms of nonlinear activation functions . The nonlinear forms proposed forperforming the functional representation are associated with a new class ofparametric neural transfer functions called rectified power sigmoid units .…

## GRADE AO Towards Near Optimal Spatially Coupled Codes With High Memories

Spatially-coupled (SC) codes are known for their threshold saturation phenomenon and low-latency windowed decoding algorithms . They also find application in various data storage systems because of their excellent performance . SC codes are constructed by partitioning an underlying block code, followed by rearranging andconcatenating the partitioned components in a “convolutional” manner .…

## From Model driven to Data driven A Survey on Active Deep Learning

Active Deep Learning (ADL) only if its predictor is deep model, where the basic learner is called as predictorand the labeling schemes is called selector . With the development of deeplearning, the selector in ADL also is experiencing the stage from model-driven to data-driven .…

## Approximate Integrals Over Bounded Volumes with Smooth Boundaries

A Radial Basis Function Generated Finite-Differences (RBF-FD) inspiredtechnique for evaluating definite integrals over bounded volumes that havesmooth boundaries in three dimensions is described . The proposed algorithm computes quadrature weights for $N$arbitrarily scattered nodes in only $O(N\mbox{ log}N)$ operations with highorders of accuracy .…

## Asymptotic Assessment of Distribution Voltage Profile Using a Nonlinear ODE Model

This paper addresses the assessment problem in aframework of nonlinear differential equations . It provides a mathematically-rigor andquantitative method for assessing how the charging/discharging of EVs affectsthe spatial profile of distribution voltage . Effectiveness of the asymptotic charcterisation of solutions of the problem is established with simulations of both simple and practicalconfigurations of the power distribution grid .…

## A Review of Graph Neural Networks and Their Applications in Power Systems

Deep neural networks have revolutionized many machine learning tasks in powersystems . Increasing number of applications in power systems where data are collected from non-Euclidean domains . Complexity of graph-structured data has brought significant challenges to the existing deep neural networks defined in Euclidean domain .…

## Learning From Revisions Quality Assessment of Claims in Argumentation at Scale

Assessing the quality of arguments and of the claims the arguments are composed of has become a key task in computational argumentation . However, even if different claims share the same stance on the same topic, their assessment depends on the prior perception and weighting of the different aspects of the topic being discussed .…

## MadDog A Web based System for Acronym Identification and Disambiguation

Acronyms and abbreviations are the short-form of longer phrases . They provide challenges for understanding the text especially if the acronym is not defined in the text or if it is used far from its definition in long texts . The web-based system is publicly available at http://iqcs.uoregon.edu:5000…

## Performance Evaluation of Convolutional Neural Networks for Gait Recognition

In this paper, a performance evaluation of well-known deep learning models ingait recognition is presented . The transfer learning scheme is adopted to pre-trained models in order to fit the models to the CASIA-Bdataset for solving a gait recognition task .…

## A new approach to extracting coronary arteries and detecting stenosis in invasive coronary angiograms

In stable coronary artery disease, reduction in mortality and/ormyocardial infarction with revascularization over medical therapy has not beenreliably achieved . We aim to develop an automatic algorithm by deep learning to extractcoronary arteries from ICAs . After segmentation, an arterial stenosis detection algorithm was developed to extract vascular centerlines and calculate arterial diameter to evaluate stenotic level .…

## Estimating the Total Volume of Queries to a Search Engine

We study the problem of estimating the total number of searches (volume) ofqueries in a specific domain, which were submitted to a search engine in agiven time period . Our statistical model assumes that the distribution ofsearches follows a Zipf’s law, and that the observed sample volumes are biasedaccordingly to three possible scenarios .…

## Blind Diagnosis for Millimeter wave Massive MIMO Systems

Millimeter-wave (mmWave) massive multiple-input multiple- input multiple-output (MIMO) systems rely on large-scale antenna arrays to combat large path-loss at mmWaveband . Current diagnostic techniques require full or partial knowledge of channel state information (CSI), which can be challenging to acquire in the presence of antenna failures .…

## Domain Dependent Speaker Diarization for the Third DIHARD Challenge

This report presents the system developed by the ABSP Laboratory team for the third DIHARD speech diarization challenge . The performance substantially improved over that of thebaseline when we optimized the thresholds for agglomerative hierarchical clustering and the parameters for dimensionality reduction during scoring for individual acoustic domains .…

## A two species micro macro model of wormlike micellar solutions and its maximum entropy closure approximations An energetic variational approach

Wormlike micelles are self-assemblies of polymer chains that can break and reform reversibly . The model incorporates a breaking and reforming process into the classical micro-macro dumbbell model forpolymeric fluids in a unified variational framework . By imposingproper dissipation in the coarse-grained level, the closure model, obtained by”closure-then-approximation”, preserves the thermodynamical structure of bothmechanical and chemical parts of the original system .…

## Active Attack Detection and Control in Constrained Cyber Physical Systems Under Prevented Actuation Attack

This paper proposes an active attack detection scheme for constrained cyber-physical systems . The proposed scheme consists of two units: 1) detection, and 2) control . The detection unit includes a set of parallel detectors, which are designed based on themultiple-model adaptive estimation approach to detect the attack and to identify the attacked actuator(s).…

## Personalization Paradox in Behavior Change Apps Lessons from a Social Comparison Based Personalized App for Physical Activity

Social comparison-based features are widely used in social computing apps . Most existing apps are not grounded in social comparison theories . This paper is among the first to automatically personalize socialcomparison targets . In the context of an m-health app for physical activity, we use artificial intelligence (AI) techniques of multi-armed bandits.…

## Process Level Representation of Scientific Protocols with Interactive Annotation

Process Execution Graphs~(PEG) is a document-level representation of real-world wet lab biochemistry protocols . We manually annotate PEGs in a corpus of complex lab protocols with a novel interactive textual simulator . We use this data to developgraph-prediction models, finding them to be good at entity identification and local relation extraction, while our corpus facilitates further exploration of long-range relations .…

## ECOL R Encouraging Copying in Novel Object Captioning with Reinforcement Learning

Novel Object Captioning is a zero-shot Image Captioning task requiringdescribing objects not seen in training captions . The ECOL-R model (Encouraging Copying ofObject Labels with Reinforced Learning) is encouraged to accurately describe the novel object labels . This isachieved via a specialised reward function in the SCST reinforcement learningframework (Rennie et al.,…

## Unsupervised Anomaly Detection and Localisation with Multi scale Interpolated Gaussian Descriptors

Current unsupervised anomaly detection and localisation systems are commonlyformulated as one-class classifiers that depend on an effective estimation of the distribution of normal images and robust criteria to identify anomalies . However, the current systems tendsto be unstable for classes that are under-represented in the training set, and the anomaly identification criteria commonly explored in thefield does not work well for multi-scale structural and non-structuralanomalies .…

## ATRM Attention based Task level Relation Module for GNN based Few shot Learning

graph neural networks (GNNs) have shown powerful ability to handle few-shot classification problem . GNNs aim to classify unseen samples whentrained with limited labeled samples per class . We propose a new relation measure method to explicitly model the task-level relation of one sample to all the others .…

## On the Performance of Image Recovery in Massive MIMO Communications

Massive MIMO (Multiple Input Multiple Output) has demonstrated as a potential candidate for 5G-and-beyond wireless networks . Alow-pass filter is exploited to enhance efficiency of the remaining noiseand artifacts reduction in the recovered image. Numerical results demonstratethe necessity of a post-filtering process in enhancing the quality of imagerecovery.…

## Particle Swarm Optimization Development of a General Purpose Optimizer

The particle swarmoptimization (PSO) method is sometimes viewed as another evolutionary algorithmbecause of their many similarities, despite not being inspired by the samemetaphor . The plain version can be programmed in a few lines of code, involving no operator design and few parameters to be tuned .…