Improving neural networks by preventing co-adaptation of feature detectors

Abstract

When a large feedforward neural network is trained on a small training set, it typically performs poorly on held-out test data. This "overfitting" is greatly reduced by randomly omitting half of the feature detectors on each training case. This prevents complex co-adaptations in which a feature detector is only helpful in the context of several other specific feature detectors. Instead, each neuron learns to detect a feature that is generally helpful for producing the correct answer given the combinatorially large variety of internal contexts in which it must operate. Random "dropout" gives big improvements on many benchmark tasks and sets new records for speech and object recognition.

Cited by in corpus (508)

• Deep Learning in Neural Networks: An Overview
• Distilling the Knowledge in a Neural Network
• Conditional Generative Adversarial Nets
• How transferable are features in deep neural networks?
• Learning Transferable Features with Deep Adaptation Networks
• Improved Regularization of Convolutional Neural Networks with Cutout
• Striving for Simplicity: The All Convolutional Net
• Attention-Based Models for Speech Recognition
• Learning Face Representation from Scratch
• Understanding Neural Networks Through Deep Visualization
• Weight Uncertainty in Neural Networks
• Convolutional Neural Networks for Sentence Classification
• Deep Learning with Limited Numerical Precision
• Convolutional Neural Network Architectures for Matching Natural Language Sentences
• DeepID3: Face Recognition with Very Deep Neural Networks
• Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
• Deep learning for molecular design - a review of the state of the art
• Deep Bayesian Active Learning with Image Data
• Training Deep Neural Networks on Noisy Labels with Bootstrapping
• Stochastic Pooling for Regularization of Deep Convolutional Neural Networks
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• On Using Monolingual Corpora in Neural Machine Translation
• A Convolutional Neural Network for Modelling Sentences
• Scalable Bayesian Optimization Using Deep Neural Networks
• End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Results
• Residual Networks of Residual Networks: Multilevel Residual Networks
• Advancements in Image Classification using Convolutional Neural Network
• SMASH: One-Shot Model Architecture Search through HyperNetworks
• Generative Moment Matching Networks
• Learning with Pseudo-Ensembles
• Learning Activation Functions to Improve Deep Neural Networks
• Compute Trends Across Three Eras of Machine Learning
• R-Drop: Regularized Dropout for Neural Networks
• Text Classification Improved by Integrating Bidirectional LSTM with Two-dimensional Max Pooling
• Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks
• Reducing Transformer Depth on Demand with Structured Dropout
• Making brain-machine interfaces robust to future neural variability
• DeepSign: Deep Learning for Automatic Malware Signature Generation and Classification
• Multi-Modal Self-Supervised Learning for Recommendation
• Star-galaxy Classification Using Deep Convolutional Neural Networks
• Effective Use of Word Order for Text Categorization with Convolutional Neural Networks
• Spatially-sparse convolutional neural networks
• Pooling Methods in Deep Neural Networks, a Review
• Regularization for Deep Learning: A Taxonomy
• Jet Flavor Classification in High-Energy Physics with Deep Neural Networks
• Winner-Take-All Autoencoders
• Less-forgetting Learning in Deep Neural Networks
• A Comprehensive guide to Bayesian Convolutional Neural Network with Variational Inference
• Dynamic Feature Integration for Simultaneous Detection of Salient Object, Edge and Skeleton
• Multi-task Neural Networks for QSAR Predictions
• An Entropy-based Pruning Method for CNN Compression
• Pyramidal Convolution: Rethinking Convolutional Neural Networks for Visual Recognition
• DART: Dropouts meet Multiple Additive Regression Trees
• Deep learning for source camera identification on mobile devices
• Deep Networks with Internal Selective Attention through Feedback Connections
• Memory Bounded Deep Convolutional Networks
• Genetic Algorithms for Evolving Deep Neural Networks
• A Comprehensive Analysis of the Role of Artificial Intelligence and Machine Learning in Modern Digital Forensics and Incident Response
• Freeze-Thaw Bayesian Optimization
• Two-Stream 3D Convolutional Neural Network for Skeleton-Based Action Recognition
• B-CNN: Branch Convolutional Neural Network for Hierarchical Classification
• Dropout Inference in Bayesian Neural Networks with Alpha-divergences
• APAC: Augmented PAttern Classification with Neural Networks
• Scale-Invariant Convolutional Neural Networks
• ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression
• Enhanced Higgs to ${\tau }^{+}{\tau }^{-}$ Searches with Deep Learning
• Recover Canonical-View Faces in the Wild with Deep Neural Networks
• Approaching the Computational Color Constancy as a Classification Problem through Deep Learning
• Deep Learning for Medical Image Segmentation
• Learning Fine-grained Image Similarity with Deep Ranking
• The Power of Sparsity in Convolutional Neural Networks
• Toxicity Prediction using Deep Learning
• Deep Structured Output Learning for Unconstrained Text Recognition
• Multi-Target Regression via Random Linear Target Combinations
• Neural network-based on-chip spectroscopy using a scalable plasmonic encoder
• On the Origin of Deep Learning
• Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
• Fisher Vectors Derived from Hybrid Gaussian-Laplacian Mixture Models for Image Annotation
• Exploring Randomly Wired Neural Networks for Image Recognition
• A Deep Learning-based Reconstruction of Cosmic Ray-induced Air Showers
• Deep Learning Classification in Asteroseismology
• Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic
• Recurrent Neural Networks for Time Series Forecasting
• FreezeOut: Accelerate Training by Progressively Freezing Layers
• LFADS - Latent Factor Analysis via Dynamical Systems
• Analyzing noise in autoencoders and deep networks
• Dataset and Neural Recurrent Sequence Labeling Model for Open-Domain Factoid Question Answering
• Revisiting Self-Supervised Visual Representation Learning
• Deep Learning with Low Precision by Half-wave Gaussian Quantization
• CL4CTR: A Contrastive Learning Framework for CTR Prediction
• DropMessage: Unifying Random Dropping for Graph Neural Networks
• A Probabilistic Theory of Deep Learning
• Mixture of Counting CNNs: Adaptive Integration of CNNs Specialized to Specific Appearance for Crowd Counting
• Neural Machine Translation with Reconstruction
• Analysis and Optimization of Convolutional Neural Network Architectures
• $\mathcal{G}$-softmax: Improving Intra-class Compactness and Inter-class Separability of Features
• RDP-Net: Region Detail Preserving Network for Change Detection
• Multiscale Vision Transformers
• OSLNet: Deep Small-Sample Classification with an Orthogonal Softmax Layer
• Raiders of the Lost Architecture: Kernels for Bayesian Optimization in Conditional Parameter Spaces
• Big Neural Networks Waste Capacity
• Visual Causal Feature Learning
• ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfer
• Heterogeneous Multi-task Learning for Human Pose Estimation with Deep Convolutional Neural Network
• Regularization and Optimization strategies in Deep Convolutional Neural Network
• Deep Temporal Appearance-Geometry Network for Facial Expression Recognition
• Improving Classifier Training Efficiency for Automatic Cyberbullying Detection with Feature Density
• Progressive Pose Attention Transfer for Person Image Generation
• A Selective Overview of Deep Learning
• EmoNets: Multimodal deep learning approaches for emotion recognition in video
• When Face Recognition Meets with Deep Learning: an Evaluation of Convolutional Neural Networks for Face Recognition
• Bayesian Neural Networks for Genetic Association Studies of Complex Disease
• Techniques for Learning Binary Stochastic Feedforward Neural Networks
• Speaker Diarization using Deep Recurrent Convolutional Neural Networks for Speaker Embeddings
• PatchNet: Hierarchical Deep Learning-Based Stable Patch Identification for the Linux Kernel
• On the Robustness of Convolutional Neural Networks to Internal Architecture and Weight Perturbations
• Learning to Discover Efficient Mathematical Identities
• TensorSCONE: A Secure TensorFlow Framework using Intel SGX
• A Novel Physics-Regularized Interpretable Machine Learning Model for Grain Growth
• A Bayesian encourages dropout
• Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks
• Improved Training of Physics-Informed Neural Networks with Model Ensembles
• Discriminative Recurrent Sparse Auto-Encoders
• On Vectorization of Deep Convolutional Neural Networks for Vision Tasks
• Online Linear Optimization via Smoothing
• Hypothesis Testing and Machine Learning: Interpreting Variable Effects in Deep Artificial Neural Networks using Cohen's f2
• Deep Models for Engagement Assessment With Scarce Label Information
• Exploring Cross-Domain Pretrained Model for Hyperspectral Image Classification
• Exploiting Noise as a Resource for Computation and Learning in Spiking Neural Networks
• Towards Understanding Label Smoothing
• KLIFF: A framework to develop physics-based and machine learning interatomic potentials
• Efficient Object Localization Using Convolutional Networks
• CASTLE: Regularization via Auxiliary Causal Graph Discovery
• Exponentially Increasing the Capacity-to-Computation Ratio for Conditional Computation in Deep Learning
• Is Joint Training Better for Deep Auto-Encoders?
• Simple vs complex temporal recurrences for video saliency prediction
• To Drop or Not to Drop: Robustness, Consistency and Differential Privacy Properties of Dropout
• Online and Offline Reinforcement Learning by Planning with a Learned Model
• Deep Learning for Identifying Iran's Cultural Heritage Buildings in Need of Conservation Using Image Classification and Grad-CAM
• Improved Residual Networks for Image and Video Recognition
• DropNeuron: Simplifying the Structure of Deep Neural Networks
• Answer Sequence Learning with Neural Networks for Answer Selection in Community Question Answering
• Solving Optimization Problems through Fully Convolutional Networks: an Application to the Travelling Salesman Problem
• Interactive Attention for Neural Machine Translation
• Joint Training of Deep Boltzmann Machines
• Optimized Three Deep Learning Models Based-PSO Hyperparameters for Beijing PM2.5 Prediction
• Deep Regression for Face Alignment
• Learning Uncertain Convolutional Features for Accurate Saliency Detection
• Project Hephaistos - II. Dyson sphere candidates from Gaia DR3, 2MASS, and WISE
• Task Augmentation by Rotating for Meta-Learning
• Catastrophic forgetting: still a problem for DNNs
• Randomness In Neural Network Training: Characterizing The Impact of Tooling
• Hierarchical learning for DNN-based acoustic scene classification
• High Performance Offline Handwritten Chinese Character Recognition Using GoogLeNet and Directional Feature Maps
• Unsupervised Deep Haar Scattering on Graphs
• Altitude Training: Strong Bounds for Single-Layer Dropout
• Sequence Model Design for Code Completion in the Modern IDE
• Dropout as a Low-Rank Regularizer for Matrix Factorization
• Towards Optimally Weighted Physics-Informed Neural Networks in Ocean Modelling
• Learning unbiased features
• Transition Propagation Graph Neural Networks for Temporal Networks
• Multi-Level Matching and Aggregation Network for Few-Shot Relation Classification
• Learnable Bernoulli Dropout for Bayesian Deep Learning
• Temporal Aggregation and Propagation Graph Neural Networks for Dynamic Representation
• Comparing Rule-Based and Deep Learning Models for Patient Phenotyping
• Neural Decision Trees
• An Attentive Sequence Model for Adverse Drug Event Extraction from Biomedical Text
• Multi-path Convolutional Neural Networks for Complex Image Classification
• Object Recognition Using Deep Neural Networks: A Survey
• Twitter Bot Detection Using Bidirectional Long Short-term Memory Neural Networks and Word Embeddings
• SimNets: A Generalization of Convolutional Networks
• Application of Deep Neural Networks to assess corporate Credit Rating
• Image Based Identification of Ghanaian Timbers Using the XyloTron: Opportunities, Risks and Challenges
• Machine learning surrogates for efficient hydrologic modeling: Insights from stochastic simulations of managed aquifer recharge
• Deep Recurrent Models with Fast-Forward Connections for Neural Machine Translation
• A PCA-Based Convolutional Network
• Machine learning and AI research for Patient Benefit: 20 Critical Questions on Transparency, Replicability, Ethics and Effectiveness
• A Bayesian Deep Learning Approach to Near-Term Climate Prediction
• Estimating Cosmological Parameters from the Dark Matter Distribution
• Bayesian Graph Neural Networks with Adaptive Connection Sampling
• Half-CNN: A General Framework for Whole-Image Regression
• End-to-end Convolutional Network for Saliency Prediction
• Treelogy: A Novel Tree Classifier Utilizing Deep and Hand-crafted Representations
• GaborNet: Gabor filters with learnable parameters in deep convolutional neural networks
• Knowledge Projection for Deep Neural Networks
• Funnel Activation for Visual Recognition
• Enhancing Rumor Detection in Social Media Using Dynamic Propagation Structures
• Referenceless Quality Estimation for Natural Language Generation
• Multi-Label Classification with Label Graph Superimposing
• Dropout Training for Support Vector Machines
• Variational Inference with Hamiltonian Monte Carlo
• Exploiting Sentence and Context Representations in Deep Neural Models for Spoken Language Understanding
• Exploring the Landscape of Ubiquitous In-home Health Monitoring: A Comprehensive Survey
• The Winning Solution to the IEEE CIG 2017 Game Data Mining Competition
• Combining the Best of Graphical Models and ConvNets for Semantic Segmentation
• A graph-based probabilistic geometric deep learning framework with online enforcement of physical constraints to predict the criticality of defects in porous materials
• Understanding Locally Competitive Networks
• Dropout with Expectation-linear Regularization
• Machine Vision for Natural Gas Methane Emissions Detection Using an Infrared Camera
• Task-Oriented Learning of Word Embeddings for Semantic Relation Classification
• On Training Implicit Models
• LocalDrop: A Hybrid Regularization for Deep Neural Networks
• Progressive Self-Supervised Attention Learning for Aspect-Level Sentiment Analysis
• Artificial Intelligence Model for Tumoral Clinical Decision Support Systems
• Prototype-based Neural Network Layers: Incorporating Vector Quantization
• Teaching Deep Convolutional Neural Networks to Play Go
• Learning to Remember Translation History with a Continuous Cache
• Generic Object Detection With Dense Neural Patterns and Regionlets
• Deep Learning Optimized Sparse Antenna Activation for Reconfigurable Intelligent Surface Assisted Communication
• Accenture at CheckThat! 2020: If you say so: Post-hoc fact-checking of claims using transformer-based models
• Can Taxonomy Help? Improving Semantic Question Matching using Question Taxonomy
• Towards Learning Convolutions from Scratch
• Arabic Multi-Dialect Segmentation: bi-LSTM-CRF vs. SVM
• Sequential Deep Learning for Credit Risk Monitoring with Tabular Financial Data
• Implementation of Deep Convolutional Neural Network in Multi-class Categorical Image Classification
• Contextual Dropout: An Efficient Sample-Dependent Dropout Module
• Deep learning for subgrid-scale turbulence modeling in large-eddy simulations of the atmospheric boundary layer
• Deep Learning Neural Networks for Emotion Classification from Text: Enhanced Leaky Rectified Linear Unit Activation and Weighted Loss
• Achieving Guidance in Applied Machine Learning through Software Engineering Techniques
• DBTagger: Multi-Task Learning for Keyword Mapping in NLIDBs Using Bi-Directional Recurrent Neural Networks
• Fashion Retail: Forecasting Demand for New Items
• Abstract Learning via Demodulation in a Deep Neural Network
• Temporal Autoencoding Restricted Boltzmann Machine
• Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
• Span-Based Constituency Parsing with a Structure-Label System and Provably Optimal Dynamic Oracles
• Leveraging Uncertainty for Improved Static Malware Detection Under Extreme False Positive Constraints
• OCoR: An Overlapping-Aware Code Retriever
• Improved Deep Convolutional Neural Network For Online Handwritten Chinese Character Recognition using Domain-Specific Knowledge
• SIDE-real: Supernova Ia Dust Extinction with truncated marginal neural ratio estimation applied to real data
• Which Contrast Does Matter? Towards a Deep Understanding of MR Contrast using Collaborative GAN
• Character-level Chinese Writer Identification using Path Signature Feature, DropStroke and Deep CNN
• On Dropout and Nuclear Norm Regularization
• Automatic Fused Multimodal Deep Learning for Plant Identification
• A Simple LSTM model for Transition-based Dependency Parsing
• SimCSE++: Improving Contrastive Learning for Sentence Embeddings from Two Perspectives
• Reducing the dilution: An analysis of the information sensitiveness of capsule network with a practical improvement method
• Surface Defect Classification in Real-Time Using Convolutional Neural Networks
• Parton distribution functions
• Structured Pruning for Efficient ConvNets via Incremental Regularization
• Deep Learning the Indus Script
• UAV-assisted Distributed Learning for Environmental Monitoring in Rural Environments
• Deep Recurrent Neural Network for Mobile Human Activity Recognition with High Throughput
• Transformed ${\ell }_1$ Regularization for Learning Sparse Deep Neural Networks
• Enhancing Scientific Papers Summarization with Citation Graph
• Event Selection and Background Rejection in Time Projection Chambers Using Convolutional Neural Networks and a Specific Application to the AdEPT Gamma-ray Polarimeter Mission
• Data Stream Stabilization for Optical Coherence Tomography Volumetric Scanning
• CIFAR-10: KNN-based Ensemble of Classifiers
• Character-Level Language Modeling with Hierarchical Recurrent Neural Networks
• The ones that got away: chemical tagging of globular cluster-origin stars with Gaia BP/RP spectra
• Context Aware Machine Learning
• Demand Forecasting using Long Short-Term Memory Neural Networks
• Hidden Classification Layers: Enhancing linear separability between classes in neural networks layers
• Word and Document Embeddings based on Neural Network Approaches
• A Scale Mixture Perspective of Multiplicative Noise in Neural Networks
• Learning Phrase Embeddings from Paraphrases with GRUs
• FR-Detect: A Multi-Modal Framework for Early Fake News Detection on Social Media Using Publishers Features
• Dependency Sensitive Convolutional Neural Networks for Modeling Sentences and Documents
• Measuring Dependence with Matrix-based Entropy Functional
• Bayesian Neural Networks at Scale: A Performance Analysis and Pruning Study
• Merging External Bilingual Pairs into Neural Machine Translation
• Catch-Up Mix: Catch-Up Class for Struggling Filters in CNN
• The Three Hundred Project: Mapping The Matter Distribution in Galaxy Clusters Via Deep Learning from Multiview Simulated Observations
• Bidirectional LSTM-CRF Attention-based Model for Chinese Word Segmentation
• On the Inference Calibration of Neural Machine Translation
• Densely Connected Recurrent Residual (Dense R2UNet) Convolutional Neural Network for Segmentation of Lung CT Images
• Automatic Assessment of the Design Quality of Python Programs with Personalized Feedback
• Piecewise Linear Multilayer Perceptrons and Dropout
• HEp-2 Cell Image Classification with Deep Convolutional Neural Networks
• Exploiting the Statistics of Learning and Inference
• Scalable and Incremental Learning of Gaussian Mixture Models
• Jointly Trained Sequential Labeling and Classification by Sparse Attention Neural Networks
• A Cross-Domain Transferable Neural Coherence Model
• Divide and Conquer: A Deep CASA Approach to Talker-independent Monaural Speaker Separation
• Radar Emitter Classification with Attribute-specific Recurrent Neural Networks
• Automatic Post-Editing for Machine Translation
• NLS: an accurate and yet easy-to-interpret regression method
• Role of Orthogonality Constraints in Improving Properties of Deep Networks for Image Classification
• Connecting the Dots: Detecting Adversarial Perturbations Using Context Inconsistency
• Light-Weighted CNN for Text Classification
• Autocorrect in the Process of Translation -- Multi-task Learning Improves Dialogue Machine Translation
• Progressive and Aligned Pose Attention Transfer for Person Image Generation
• Scene Change Detection Using Multiscale Cascade Residual Convolutional Neural Networks
• Automatic selection of active spaces for strongly correlated systems using machine learning algorithms
• Improving Neural Network Generalization by Combining Parallel Circuits with Dropout
• Continual Learning for Natural Language Generation in Task-oriented Dialog Systems
• Improving random walk rankings with feature selection and imputation
• LDMNet: Low Dimensional Manifold Regularized Neural Networks
• Knowledge Graph Embedding with Atrous Convolution and Residual Learning
• Learning Optimal Data Augmentation Policies via Bayesian Optimization for Image Classification Tasks
• Thoughts on a Recursive Classifier Graph: a Multiclass Network for Deep Object Recognition
• Automatic Extraction of Secrets from the Transistor Jungle using Laser-Assisted Side-Channel Attacks
• The Benefit of the Doubt: Uncertainty Aware Sensing for Edge Computing Platforms
• $N^4$-Fields: Neural Network Nearest Neighbor Fields for Image Transforms
• Gaussian Process Neurons Learn Stochastic Activation Functions
• Neural Network Regularization via Robust Weight Factorization
• Evaluating the Rainbow DQN Agent in Hanabi with Unseen Partners
• Open DNN Box by Power Side-Channel Attack
• An Intelligent Approach to Detecting Novel Fault Classes for Centrifugal Pumps Based on Deep CNNs and Unsupervised Methods
• Machine Learning Automation Toolbox (MLaut)
• Multi-Loss Sub-Ensembles for Accurate Classification with Uncertainty Estimation
• A Study of Gender Classification Techniques Based on Iris Images: A Deep Survey and Analysis
• Autoencoder Regularized Network For Driving Style Representation Learning
• MHAttnSurv: Multi-Head Attention for Survival Prediction Using Whole-Slide Pathology Images
• DropSample: A New Training Method to Enhance Deep Convolutional Neural Networks for Large-Scale Unconstrained Handwritten Chinese Character Recognition
• Learning-accelerated A* Search for Risk-aware Path Planning
• WeightNet: Revisiting the Design Space of Weight Networks
• PhytNet -- Tailored Convolutional Neural Networks for Custom Botanical Data
• ADER: Adaptively Distilled Exemplar Replay Towards Continual Learning for Session-based Recommendation
• Robust Deep Graph Based Learning for Binary Classification
• Generating Person Images with Appearance-aware Pose Stylizer
• Can deep learning help you find the perfect match?
• Combine Convolution with Recurrent Networks for Text Classification
• Branch-Cooperative OSNet for Person Re-Identification
• Financial Time Series Prediction Using Deep Learning
• Cross-relation Cross-bag Attention for Distantly-supervised Relation Extraction
• Learning Machines Implemented on Non-Deterministic Hardware
• Defending Adversarial Attacks by Correcting logits
• Maxmin convolutional neural networks for image classification
• Learning Compact Convolutional Neural Networks with Nested Dropout
• Self-Orthogonality Module: A Network Architecture Plug-in for Learning Orthogonal Filters
• Differentially Private Variational Dropout
• Locally Smoothed Neural Networks
• Less-forgetful Learning for Domain Expansion in Deep Neural Networks
• Coloring the Black Box: Visualizing neural network behavior with a self-introspective model
• On the Equivalence Between Deep NADE and Generative Stochastic Networks
• Safety of Quark/Gluon Jet Classification
• A novel combination of theoretical analysis and data-driven method for reconstruction of structural defects
• Investigating the Relationship Between Dropout Regularization and Model Complexity in Neural Networks
• A new way to constrain the densities of intra-group medium in groups of galaxies with convolutional neural networks
• Research Progress of Convolutional Neural Network and its Application in Object Detection
• Abusive Language Detection and Characterization of Twitter Behavior
• A Deep Neural Architecture for Sentence-level Sentiment Classification in Twitter Social Networking
• Multiple Range-Restricted Bidirectional Gated Recurrent Units with Attention for Relation Classification
• FLFE: A Communication-Efficient and Privacy-Preserving Federated Feature Engineering Framework
• Deconfusing intensity maps with neural networks
• Fine-tuning Handwriting Recognition systems with Temporal Dropout
• Training like Playing: A Reinforcement Learning And Knowledge Graph-based framework for building Automatic Consultation System in Medical Field
• Regularizing Recurrent Networks - On Injected Noise and Norm-based Methods
• Analysis of skin lesion images with deep learning
• Differential equations as models of deep neural networks
• Clothing Retrieval with Visual Attention Model
• Learning Light-Weight Translation Models from Deep Transformer
• Solving inverse problems with deep neural networks driven by sparse signal decomposition in a physics-based dictionary
• Cross-Media Similarity Evaluation for Web Image Retrieval in the Wild
• MUSCLE: Strengthening Semi-Supervised Learning Via Concurrent Unsupervised Learning Using Mutual Information Maximization
• Uncertainty Estimation and Calibration with Finite-State Probabilistic RNNs
• Deep Data Flow Analysis
• Inherent Weight Normalization in Stochastic Neural Networks
• CrescendoNet: A Simple Deep Convolutional Neural Network with Ensemble Behavior
• Table-Of-Contents generation on contemporary documents
• Image aesthetic evaluation using paralleled deep convolution neural network
• Beyond Dropout: Feature Map Distortion to Regularize Deep Neural Networks
• Variational inference of latent state sequences using Recurrent Networks
• Uncertainty Quantification of Locally Nonlinear Dynamical Systems using Neural Networks
• GSNs : Generative Stochastic Networks
• Who Blames Whom in a Crisis? Detecting Blame Ties from News Articles Using Neural Networks
• Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models
• Predicting TED Talk Ratings from Language and Prosody
• Exploiting Spatio-Temporal Structure with Recurrent Winner-Take-All Networks
• The Earth Mover's Pinball Loss: Quantiles for Histogram-Valued Regression
• Machine Learning Approach to Earthquake Rupture Dynamics
• Weakly supervised learning for pattern classification in serial femtosecond crystallography
• Entity Candidate Network for Whole-Aware Named Entity Recognition
• A simple and effective postprocessing method for image classification
• Neuron with Steady Response Leads to Better Generalization
• Informed Machine Learning, Centrality, CNN, Relevant Document Detection, Repatriation of Indigenous Human Remains
• Recycle deep features for better object detection
• Unsupervised detection and open-set classification of fast-ramped flexibility activation events
• What training reveals about neural network complexity
• Pre-interpolation loss behaviour in neural networks
• A Multi-Level Attention Model for Evidence-Based Fact Checking
• Routing Towards Discriminative Power of Class Capsules
• Delving Deep into Liver Focal Lesion Detection: A Preliminary Study
• Classification of Breast Cancer Lesions in Ultrasound Images by using Attention Layer and loss Ensembles in Deep Convolutional Neural Networks
• Towards Controlled and Diverse Generation of Article Comments
• Investigation of Training Label Error Impact on RNN-T
• Automatic Whole-body Bone Age Assessment Using Deep Hierarchical Features
• Using multi-task learning to improve the performance of acoustic-to-word and conventional hybrid models
• Leveraging Newswire Treebanks for Parsing Conversational Data with Argument Scrambling
• Fused Deep Convolutional Neural Network for Precision Diagnosis of COVID-19 Using Chest X-Ray Images
• A Capsule-unified Framework of Deep Neural Networks for Graphical Programming
• A Multilingual Encoding Method for Text Classification and Dialect Identification Using Convolutional Neural Network
• Zero-Shot Human-Object Interaction Recognition via Affordance Graphs
• Learning Shared Encoding Representation for End-to-End Speech Recognition Models
• Benchmarking Approximate Inference Methods for Neural Structured Prediction
• Neural Neighborhood Encoding for Classification
• Adversarial Robustness for Machine Learning Cyber Defenses Using Log Data
• Monte Carlo dropout increases model repeatability
• Adma: A Flexible Loss Function for Neural Networks
• Stochastic Bottleneck: Rateless Auto-Encoder for Flexible Dimensionality Reduction
• Adaptive Weighting Depth-variant Deconvolution of Fluorescence Microscopy Images with Convolutional Neural Network
• Learning in Deep Neural Networks Using a Biologically Inspired Optimizer
• Optical Convolutional Neural Networks -- Combining Silicon Photonics and Fourier Optics for Computer Vision
• A Weakly Supervised Approach to Train Temporal Relation Classifiers and Acquire Regular Event Pairs Simultaneously
• Tikhonov Regularization for Long Short-Term Memory Networks
• Training Image Selection using Recurrent Neural Networks: An Application in Hydrogeology
• Is the User Enjoying the Conversation? A Case Study on the Impact on the Reward Function
• Construction of a far ultraviolet all sky map from an incomplete survey: Application of a deep learning algorithm
• Comparing methods addressing multi-collinearity when developing prediction models
• Cascade Attentive Dropout for Weakly Supervised Object Detection
• Closing the loop between neural network simulators and the OpenAI Gym
• Generalized Constraints as A New Mathematical Problem in Artificial Intelligence: A Review and Perspective
• Rethinking the Value of Transformer Components
• Disturbing Target Values for Neural Network Regularization
• The Surprising Effectiveness of Visual Odometry Techniques for Embodied PointGoal Navigation
• Learning event representation: As sparse as possible, but not sparser
• A Joint and Domain-Adaptive Approach to Spoken Language Understanding
• Semi-supervised Learning for Marked Temporal Point Processes
• A Template-guided Hybrid Pointer Network for Knowledge-basedTask-oriented Dialogue Systems
• SCONE: Supernova Classification with a Convolutional Neural Network
• Livewired Neural Networks: Making Neurons That Fire Together Wire Together
• A Package for the Automated Classification of Images Containing Supernova Light Echoes
• A Transformer-based Autoregressive Decoder Architecture for Hierarchical Text Classification
• Connections Between Pairs of Filters Improve the Accuracy of Convolutional Neural Networks
• Machine-learning-enabled characterization of individual ring resonators in integrated photonic lattices
• A Multichannel Convolutional Neural Network For Cross-language Dialog State Tracking
• Unsupervised End-to-End Training with a Self-Defined Target
• Deep Collaborative Learning for Visual Recognition
• PULASki: Learning inter-rater variability using statistical distances to improve probabilistic segmentation
• Deep set conditioned latent representations for action recognition
• Lexical Translation Model Using a Deep Neural Network Architecture
• Advances in Collaborative Filtering and Ranking
• Latent Opinions Transfer Network for Target-Oriented Opinion Words Extraction
• Conditional Generative Adversarial Networks for Optimal Path Planning
• Towards Lower-Dose PET using Physics-Based Uncertainty-Aware Multimodal Learning with Robustness to Out-of-Distribution Data
• End-to-End Chinese Parsing Exploiting Lexicons
• Quaternion-Based Graph Convolution Network for Recommendation
• Dropout Regularization for Self-Supervised Learning of Transformer Encoder Speech Representation
• Generalization by design: Shortcuts to Generalization in Deep Learning
• Structured in Space, Randomized in Time: Leveraging Dropout in RNNs for Efficient Training
• Using Context Events in Neural Network Models for Event Temporal Status Identification
• Fast Landmark Localization with 3D Component Reconstruction and CNN for Cross-Pose Recognition
• Topical Behavior Prediction from Massive Logs
• CODA: Constructivism Learning for Instance-Dependent Dropout Architecture Construction
• Open-domain Topic Identification of Out-of-domain Utterances using Wikipedia
• Quantal synaptic dilution enhances sparse encoding and dropout regularisation in deep networks
• Sparse, guided feature connections in an Abstract Deep Network
• Regularization in ResNet with Stochastic Depth
• RoutingGAN: Routing Age Progression and Regression with Disentangled Learning
• Enhanced Aspect-Based Sentiment Analysis Models with Progressive Self-supervised Attention Learning
• Efficient Sparse Artificial Neural Networks
• Back to Square One: Superhuman Performance in Chutes and Ladders Through Deep Neural Networks and Tree Search
• Beyond Short Clips: End-to-End Video-Level Learning with Collaborative Memories
• Estimating air quality co-benefits of energy transition using machine learning
• Targeted Deep Learning: Framework, Methods, and Applications
• Measuring Fine-Grained Domain Relevance of Terms: A Hierarchical Core-Fringe Approach
• FINNger -- Applying artificial intelligence to ease math learning for children
• Pseudo-positive regularization for deep person re-identification
• Data-driven discovery of interpretable causal relations for deep learning material laws with uncertainty propagation
• Block-Cyclic Stochastic Coordinate Descent for Deep Neural Networks
• Regularizing Neural Networks via Stochastic Branch Layers
• Graph Interpolating Activation Improves Both Natural and Robust Accuracies in Data-Efficient Deep Learning
• Development of digitally obtainable 10-year risk scores for depression and anxiety in the general population
• Optimized Sharing of Coefficients in Parallel Filter Banks
• A Distributed Deep Representation Learning Model for Big Image Data Classification
• Geometric Neural Phrase Pooling: Modeling the Spatial Co-occurrence of Neurons
• Deep Transform: Time-Domain Audio Error Correction via Probabilistic Re-Synthesis
• Crossover-Net: Leveraging the Vertical-Horizontal Crossover Relation for Robust Segmentation
• Horn: A System for Parallel Training and Regularizing of Large-Scale Neural Networks
• Predicting Gene Expression Between Species with Neural Networks
• Towards Robust Classification with Image Quality Assessment
• AMC-Loss: Angular Margin Contrastive Loss for Improved Explainability in Image Classification
• Mixture separability loss in a deep convolutional network for image classification
• What needles do sparse neural networks find in nonlinear haystacks
• Sequential Feature Filtering Classifier
• Deep Learning-based Stress Determinator for Mouse Psychiatric Analysis using Hippocampus Activity
• C2G-Net: Exploiting Morphological Properties for Image Classification
• Invexifying Regularization of Non-Linear Least-Squares Problems
• Randomness and Interpolation Improve Gradient Descent
• Adaptive Hierarchical Decomposition of Large Deep Networks
• Analysis of Video Feature Learning in Two-Stream CNNs on the Example of Zebrafish Swim Bout Classification
• Think Global, Act Local: Relating DNN generalisation and node-level SNR
• Interpretable Deep Learning for Stock Returns: A Consensus-Bottleneck Asset Pricing Model
• Predicting Toxicity from Gene Expression with Neural Networks
• Search Intelligence: Deep Learning For Dominant Category Prediction
• Use Generalized Representations, But Do Not Forget Surface Features
• A deep learning approach to the structural analysis of proteins
• RotationOut as a Regularization Method for Neural Network
• Diabetic Retinopathy detection by retinal image recognizing
• Using Context Information to Enhance Simple Question Answering
• DropRegion Training of Inception Font Network for High-Performance Chinese Font Recognition
• Supervised Learning in Temporally-Coded Spiking Neural Networks with Approximate Backpropagation
• Shakeout: A New Approach to Regularized Deep Neural Network Training
• Learning Connectivity of Neural Networks from a Topological Perspective
• Teaching a Machine to Diagnose a Heart Disease; Beginning from digitizing scanned ECGs to detecting the Brugada Syndrome (BrS)
• Deep Representation with ReLU Neural Networks
• Recognition Confidence Analysis of Handwritten Chinese Character with CNN
• Belief Flows of Robust Online Learning
• Text Classification through Glyph-aware Disentangled Character Embedding and Semantic Sub-character Augmentation
• Paradigm Shift in Language Modeling: Revisiting CNN for Modeling Sanskrit Originated Bengali and Hindi Language
• Two-Stream Appearance Transfer Network for Person Image Generation
• ST-ABN: Visual Explanation Taking into Account Spatio-temporal Information for Video Recognition
• How to boost autoencoders?
• Convergence Analysis of Homotopy-SGD for non-convex optimization
• Self-Supervision and Spatial-Sequential Attention Based Loss for Multi-Person Pose Estimation
• Multi-concept adversarial attacks
• The Long-Short Story of Movie Description
• Optimizing Neural Network for Computer Vision task in Edge Device
• Turning old models fashion again: Recycling classical CNN networks using the Lattice Transformation
• Active Learning for Argument Mining: A Practical Approach
• Image Disguise based on Generative Model
• Learning Energy-Based Approximate Inference Networks for Structured Applications in NLP
• LFGCN: Levitating over Graphs with Levy Flights
• Application of Facial Recognition using Convolutional Neural Networks for Entry Access Control
• Out-of-Distribution Example Detection in Deep Neural Networks using Distance to Modelled Embedding
• HOTCAKE: Higher Order Tucker Articulated Kernels for Deeper CNN Compression
• Only sparsity based loss function for learning representations
• Learn to Focus: Hierarchical Dynamic Copy Network for Dialogue State Tracking