Visformer The Vision friendly Transformer

Visformer outperforms both theTransformer-based and convolution-based models in terms of ImageNetclassification accuracy . The advantage becomes more significant when the training set is smaller . The code is available at https://://://github.com/danczs/Visformer. It is abbreviated from the `Vision-friendly Transformer’ with the same computational complexity as the Visformer architecture .…

Dynamic Degradation for Image Restoration and Fusion

The DDRF-Net is capable of solving twoproblems, i.e., static restoration and fusion, dynamic degradation . In order to solve the static fusion problem of existing methods, dynamic convolution is introduced . In addition, adynamic degradation kernel is proposed to improve the robustness of imagerestoration and fusion .…

One parameter family of acquisition functions for efficient global optimization

Bayesian optimization (BO) with Gaussian processes is a powerful methodology to optimize an expensive black-box function . The expected improvement (EI) and probability of improvement (PI) are among the most widely used schemes for BO . The proposed method isnumerically inexpensive, is easy to implement, can be easily parallelized, and on benchmark tasks shows a performance superior to EI and GP-UCB .…

Quadratic Payments with constrained probabilities

“Quadratic Payments: A Primer” is written by Vitalik Buterin . The book focuses on the voting scenario depicted in the book . It aims to create a simple referendum model with more realistic outcome probability and marginal probability qualitativeshapes . It also discusses how quadratic payments can be generalized to take into account these new functions shapes, and the way they are still quadratically is discussed .…

Optimal controller synthesis for timed systems

Weighted timed games are zero-sum games played by two players on a timedautomaton equipped with weights . One player wants to minimise thecumulative weight while reaching a target . Used in a reactive synthesisperspective, this quantitative extension of timed games allows one to measure the quality of controllers in real-time systems .…

Predicting Depressive Symptom Severity through Individuals Nearby Bluetooth Devices Count Data Collected by Mobile Phones A Preliminary Longitudinal Study

The Bluetooth sensor embedded in mobile phones provides an unobtrusive,continuous, and cost-efficient means to capture individuals’ proximityinformation . This paper aims to explore theNBDC data’s value in predicting depressive symptom severity as measured via the8-item Patient Health Questionnaire (PHQ-8) The data used in this paperincluded 2,886 bi-weekly PHQ- 8 records collected from 316 participants in the Netherlands, Spain, and the UK as part of the EU RADAR-CNS study .…

dualFace Two Stage Drawing Guidance for Freehand Portrait Sketching

dualFace consists of two-stage drawing assistance toprovide global and local visual guidance: global guidance, which helps users draw contour lines of portraits, and local guidance, inspired by traditional artist workflows in portrait drawing . In the stage of local guidance users synthesize detailed portrait images with a deep generative model, but use the synthesized results as detailed drawing guidance .…

Easy and Efficient Transformer Scalable Inference Solution For large NLP mode

EET achieves a 1.5-15x state-of-art speedup varying with context length . Easy and Efficient Transformer (EET) has a significantperformance improvement over the existing schemes . EET is available on GitHub and is available at:https://://://github.com/NetEase-FuXi/EET. The ultra-large-scale pre-training model can effectively improve the effect of a variety of tasks, and it also brings a heavy computational burden toinference.…

Diverse Image Inpainting with Bidirectional and Autoregressive Transformers

Image inpainting is an underdetermined inverse problem, it naturally allowsdiverse contents that fill up missing or corrupted regions reasonably andrealistically . Prevalent approaches using convolutional neural networks (CNNs) suffer from limited perception fields for capturing global features . We propose BAT-Fill, an image inpaining framework with a novel bidirectionalautoregressive transformer (BAT-Fill) that models deep bidsirectional contexts for the generation of diverse contents .…

Variational Pedestrian Detection

Pedestrian detection in a crowd is a challenging task due to a high number of human instances, which brings ambiguity and optimizationdifficulties to the current IoU-based ground truth assignment procedure inclassical object detection methods . The proposed algorithm serves as an efficient solution to handle the dense pedestrian detection problem for the case of single-stagedetectors .…

Mutual Contrastive Learning for Visual Representation Learning

Mutual Contrastive Learning (MCL) is a collaborative learning method for general visual representation learning . The core idea of MCL is to perform mutual interaction and transfer of contrastive distributions among acohort of models . Benefiting from MCL, each model can learn extra contrastiveknowledge from others, leading to more meaningful feature representations for recognition tasks .…

Ideology in Open Source Development

Open source development, to a great extent, is a type of social movement in which shared ideologies play critical roles . Ideology determines how participants make sense of things, shapes their thoughts, actions, and interactions . We furtherargue the imperatives of developing an empirical theory of ideology in opensource development, and propose a research agenda for developing such a theory .…

Free Vibration analysis of Curvilinearly Stiffened Composite plates with an arbitrarily shaped cutout using Isogeometric Analysis

This paper focuses on the isogeometric vibration analysis of curvilinearlystiffened composite panels . The stiffness matrices and the mass matrices are derived using the first-order shear deformation theory (FSDT) The presentmethod models the plate and the stiffener separately, which allows thestiffener element nodes to not coincide with the plate shell-element nodes .…

VCGAN Video Colorization with Hybrid Generative Adversarial Network

Hybrid GenerativeAdversarial Network (VCGAN) addresses two prevalent issues in the videocolorization domain: Temporal consistency and unification of colorizationnetwork and refinement network into a single architecture . The hybrid VCGAN strikes a good balance between color vividness and videocontinuity . To improve the consistency of farframes, we propose a dense long-term loss that smooths the temporal disparity of every two remote frames.…

An Exploration into why Output Regularization Mitigates Label Noise

Label noise presents a real challenge for supervised learning algorithms . Noise robust losses is one of the more promising approaches for dealing with label noise . But their ability to mitigate label noise lackmathematical rigor. In this work we aim at closing this gap by showing thatlosses, which incorporate an output regularization term, become symmetric .…

FedDPGAN Federated Differentially Private Generative Adversarial Networks Framework for the Detection of COVID 19 Pneumonia

Existing deep learning technologies generally learn the features of chest X-ray data generated by Generative Adversarial Networks (GAN) to diagnose COVID-19 pneumonia . However, the above methods have a critical challenge: dataprivacy . GAN will leak the semantic information of the training data which can be used to reconstruct the training samples by attackers .…

A unified Neural Network Approach to E CommerceRelevance Learning

Result relevance scoring is critical to e-commerce search user experience . Traditional information retrieval methods focus on keyword matching and counting-based numeric features . We describe a feed-forward neuralmodel to provide relevance score for (query, item) pairs . We found significantimprovement over GBDT baseline as well as several off-the-shelf deep-learningbaselines on an independently constructed ratings dataset .…

Provenance based Data Skipping TechReport

Database systems analyze queries to determine upfront which data is needed for answering them and use indexes and other physical design techniques to speed-up access to that data . For important classes of queries, e.g.,HAVING and top-k queries, it is impossible to determine up-front what data is relevant .…

Unified Spatio Temporal Modeling for Traffic Forecasting using Graph Neural Network

Research in deep learning models to forecast traffic intensities has gainedgreat attention in recent years due to their capability to capture the complexspatio-temporal relationships within the traffic data . We propose a UnifiedSpatio-Temporal Graph Convolution Network (USTGCN) for traffic forecasting that performs both spatial and temporal aggregation through direct informationpropagation across different timestamp nodes .…

Learning Aided Auctioning for Opportunistic Scheduling in a Wireless Optical Network

This paper focusses on Service Level Agreement (SLA) based end-to-end Quality of Service (QoS) maintenance across a wireless optical integrated network . The wireless network used is Long Term Evolution/Long Term Evolution Advanced(LTE/LTE-A) mobile network and the optical network is comprised of an EthernetPassive Optical Network (EPON) The proposal allows the users of the integrated network to decide the payment they want to make in order toopportunistically avail bandwidth .…

Automated Driver Testing for Small Footprint Embedded Systems

Embedded systems represent a billionaire market and are present in most of the processors produced in the world . The strong interaction between drivers and external peripherals often hamper embedded software development, in special the testing task . The proposed solution was successfully implemented andvalidated using different protocols .…

Non Parametric Few Shot Learning for Word Sense Disambiguation

Word sense disambiguation (WSD) is a long-standing problem in naturallanguage processing . A significant challenge in supervised all-words WSD is to classify among senses for a majority of words that lie in the long-taildistribution . In this work, we proposeMetricWSD, a non-parametric few-shot learning approach to mitigate this dataimbalance issue .…

Revisiting the size effect in software fault prediction models

In object oriented (OO) software systems, class size has been acknowledged as having an indirect effect on the relationship between certain characteristics, captured via metrics, and faultproneness . In particular, size appears to have a more significant mediation effect on CBO andFan-out than other metrics, although the evidence is not consistent in allexamined systems .…

Android OS CASE STUDY

Android is a mobile operating system based on a modified version of the Linux kernel and other open source software . It is an operating system forlow powered devices that run on battery and are full of hardware like GlobalPositioning System (GPS) receivers, cameras, light and orientation sensors, Wi-Fi and LTE (4G telephony) connectivity and a touch screen .…

We Haven t Gone Paperless Yet Why the Printing Press Can Help Us Understand Data and AI

This paper argues that the effects of datafication should be understood as a constitutive shift in social and political relations . We use analogy of the printing press toprovide a framework for understanding constitutive change . We highlight thattechnologies such as data fication and AI both disrupted extant power arrangements, leading to decentralization, and triggered a recentralization of power by new actors better adapted toleveraging the new technology .…

Evaluating the Values of Sources in Transfer Learning

Transfer learning that adapts a model trained on data-rich sources to low-resource targets has been widely applied in natural language processing . When training a transfer model over multiple sources, not everysource is equally useful for the target . To better transfer a model, it isessential to understand the values of the sources .…

3D TalkEmo Learning to Synthesize 3D Emotional Talking Head

3D-TalkEmo, adeep neural network that generates 3D talking head animation with variousemotions . We create a large 3D dataset with synchronized audios andvideos, rich corpus, as well as various emotion states of different persons . We propose novel 3D face representation structure -geometry map by classical multi-dimensional scaling analysis .…