Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

Abstract

We present Fashion-MNIST, a new dataset comprising of 28x28 grayscale images of 70,000 fashion products from 10 categories, with 7,000 images per category. The training set has 60,000 images and the test set has 10,000 images. Fashion-MNIST is intended to serve as a direct drop-in replacement for the original MNIST dataset for benchmarking machine learning algorithms, as it shares the same image size, data format and the structure of training and testing splits. The dataset is freely available at https://github.com/zalandoresearch/fashion-mnist

Dataset is freely available at https://github.com/zalandoresearch/fashion-mnist Benchmark is available at http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/

References in corpus (1)

• EMNIST: an extension of MNIST to handwritten letters

Cited by in corpus (525)

• Random Erasing Data Augmentation
• Prevalence of Neural Collapse during the terminal phase of deep learning training
• Agnostic Federated Learning
• Blockchain-based Federated Learning with Secure Aggregation in Trusted Execution Environment for Internet-of-Things
• Analysis and Optimal Edge Assignment For Hierarchical Federated Learning on Non-IID Data
• Misalignment Resilient Diffractive Optical Networks
• Peer-to-peer Federated Learning on Graphs
• One-Class Classification: A Survey
• Learning and Evaluating Representations for Deep One-class Classification
• Enhanced Convolutional Neural Tangent Kernels
• Auditing Differentially Private Machine Learning: How Private is Private SGD?
• On the Effectiveness of Mitigating Data Poisoning Attacks with Gradient Shaping
• Geometric Dataset Distances via Optimal Transport
• Differentially Private Learning Needs Better Features (or Much More Data)
• Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks
• Lifelong Teacher-Student Network Learning
• Mitigating Backdoor Attacks in Federated Learning
• Data Poisoning Attacks Against Federated Learning Systems
• Adversarial Examples in Modern Machine Learning: A Review
• Sample Selection with Uncertainty of Losses for Learning with Noisy Labels
• Superposition of many models into one
• Balancing Accuracy and Integrity for Reconfigurable Intelligent Surface-aided Over-the-Air Federated Learning
• Out-of-distribution Detection in Classifiers via Generation
• Hierarchically Fair Federated Learning
• Why is the Mahalanobis Distance Effective for Anomaly Detection?
• A framework for the extraction of Deep Neural Networks by leveraging public data
• Capsule Networks with Max-Min Normalization
• On the Origin of Implicit Regularization in Stochastic Gradient Descent
• Genetic Programming-Based Evolutionary Deep Learning for Data-Efficient Image Classification
• Finite Versus Infinite Neural Networks: an Empirical Study
• Detecting Out-of-Distribution Inputs in Deep Neural Networks Using an Early-Layer Output
• Measuring and Explaining the Inter-Cluster Reliability of Multidimensional Projections
• Fuzzy Pooling
• Hyper-Sphere Quantization: Communication-Efficient SGD for Federated Learning
• Tensor Normalization and Full Distribution Training
• The Lottery Tickets Hypothesis for Supervised and Self-supervised Pre-training in Computer Vision Models
• Tensor-based algorithms for image classification
• Efficient Continual Learning with Modular Networks and Task-Driven Priors
• The Usual Suspects? Reassessing Blame for VAE Posterior Collapse
• Defining Benchmarks for Continual Few-Shot Learning
• Measurements of Three-Level Hierarchical Structure in the Outliers in the Spectrum of Deepnet Hessians
• Decentralized federated learning of deep neural networks on non-iid data
• Feature Space Singularity for Out-of-Distribution Detection
• Bridging Adversarial Robustness and Gradient Interpretability
• Parameterized Knowledge Transfer for Personalized Federated Learning
• Learning Generative Models across Incomparable Spaces
• Decentralized Bayesian Learning over Graphs
• LDP-Fed: Federated Learning with Local Differential Privacy
• Increasing Liquid State Machine Performance with Edge-of-Chaos Dynamics Organized by Astrocyte-modulated Plasticity
• Regular Polytope Networks
• On the Iteration Complexity of Hypergradient Computation
• Virus-MNIST: A Benchmark Malware Dataset
• NeurIPS 2020 Competition: Predicting Generalization in Deep Learning
• Scalable Rule-Based Representation Learning for Interpretable Classification
• XNAS: Neural Architecture Search with Expert Advice
• Dual Manifold Adversarial Robustness: Defense against Lp and non-Lp Adversarial Attacks
• Detecting Out-of-Distribution Examples with In-distribution Examples and Gram Matrices
• Deep Learning Through the Lens of Example Difficulty
• Exposing Reliability Degradation and Mitigation in Approximate DNNs under Permanent Faults
• A Neural Scaling Law from the Dimension of the Data Manifold
• Distribution Density, Tails, and Outliers in Machine Learning: Metrics and Applications
• CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks
• Towards an Adversarially Robust Normalization Approach
• Unsupervised Two-Stage Anomaly Detection
• Stragglers Are Not Disaster: A Hybrid Federated Learning Algorithm with Delayed Gradients
• Anomaly Detection with Tensor Networks
• SIMDive: Approximate SIMD Soft Multiplier-Divider for FPGAs with Tunable Accuracy
• Adversarial Infidelity Learning for Model Interpretation
• Coalesced Multi-Output Tsetlin Machines with Clause Sharing
• A Gradient Boosting Approach for Training Convolutional and Deep Neural Networks
• An Asymmetric Contrastive Loss for Handling Imbalanced Datasets
• Semi-Supervised Learning by Label Gradient Alignment
• Unsupervised Out-of-Distribution Detection with Batch Normalization
• Soft-Root-Sign Activation Function
• EvoAAA: An evolutionary methodology for automated \neural autoencoder architecture search
• The Search for Sparse, Robust Neural Networks
• Accurate and Fast Federated Learning via IID and Communication-Aware Grouping
• On the Vulnerability of Capsule Networks to Adversarial Attacks
• Coverage Testing of Deep Learning Models using Dataset Characterization
• AutoOD: Automated Outlier Detection via Curiosity-guided Search and Self-imitation Learning
• Variational Autoencoders with Normalizing Flow Decoders
• Membership Model Inversion Attacks for Deep Networks
• Deep Active Learning: Unified and Principled Method for Query and Training
• On the Theory of Implicit Deep Learning: Global Convergence with Implicit Layers
• Communication-Efficient Robust Federated Learning Over Heterogeneous Datasets
• Under the Hood of Neural Networks: Characterizing Learned Representations by Functional Neuron Populations and Network Ablations
• Embedded hyper-parameter tuning by Simulated Annealing
• Latent Translation: Crossing Modalities by Bridging Generative Models
• Adversarial Detection and Correction by Matching Prediction Distributions
• Towards Understanding the Transferability of Deep Representations
• Benchmarking Quantized Neural Networks on FPGAs with FINN
• Evaluating the Robustness of Nearest Neighbor Classifiers: A Primal-Dual Perspective
• FathomNet: An underwater image training database for ocean exploration and discovery
• Variational Laplace Autoencoders
• Interpretability is a Kind of Safety: An Interpreter-based Ensemble for Adversary Defense
• Generalization bounds for deep learning
• Capsule networks with non-iterative cluster routing
• Can Single Neurons Solve MNIST? The Computational Power of Biological Dendritic Trees
• Detecting Adversarial Examples and Other Misclassifications in Neural Networks by Introspection
• Image reconstruction through a multimode fiber with a simple neural network architecture
• The Hidden Uncertainty in a Neural Networks Activations
• Uncertainty Estimation and Out-of-Distribution Detection for Counterfactual Explanations: Pitfalls and Solutions
• Towards Self-Adaptive Metric Learning On the Fly
• PuVAE: A Variational Autoencoder to Purify Adversarial Examples
• MMFashion: An Open-Source Toolbox for Visual Fashion Analysis
• Bayesian Learning of Neural Network Architectures
• On-FPGA Training with Ultra Memory Reduction: A Low-Precision Tensor Method
• Federated Learning System without Model Sharing through Integration of Dimensional Reduced Data Representations
• SemiNLL: A Framework of Noisy-Label Learning by Semi-Supervised Learning
• Improved Bilevel Model: Fast and Optimal Algorithm with Theoretical Guarantee
• State-Regularized Recurrent Neural Networks
• Are all outliers alike? On Understanding the Diversity of Outliers for Detecting OODs
• Why Generalization in RL is Difficult: Epistemic POMDPs and Implicit Partial Observability
• Alleviating Mode Collapse in GAN via Diversity Penalty Module
• Annealing Genetic GAN for Minority Oversampling
• Pixel-wise Conditioned Generative Adversarial Networks for Image Synthesis and Completion
• Deep Density-based Image Clustering
• Stateful Detection of Model Extraction Attacks
• One-pixel Signature: Characterizing CNN Models for Backdoor Detection
• Delta-STN: Efficient Bilevel Optimization for Neural Networks using Structured Response Jacobians
• On the Necessity and Effectiveness of Learning the Prior of Variational Auto-Encoder
• Fast Inference in Capsule Networks Using Accumulated Routing Coefficients
• RingFed: Reducing Communication Costs in Federated Learning on Non-IID Data
• Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks
• Soft Gradient Boosting Machine
• Detecting OODs as datapoints with High Uncertainty
• Model Weight Theft With Just Noise Inputs: The Curious Case of the Petulant Attacker
• AdvKnn: Adversarial Attacks On K-Nearest Neighbor Classifiers With Approximate Gradients
• Invert and Defend: Model-based Approximate Inversion of Generative Adversarial Networks for Secure Inference
• The Geometry of Sign Gradient Descent
• SwiftNet: Using Graph Propagation as Meta-knowledge to Search Highly Representative Neural Architectures
• Impact of Feedback Type on Explanatory Interactive Learning
• Bayesian Inference with Certifiable Adversarial Robustness
• Pushing the Limits of Capsule Networks
• Fix Your Features: Stationary and Maximally Discriminative Embeddings using Regular Polytope (Fixed Classifier) Networks
• Deep Image Clustering with Category-Style Representation
• Towards Characterizing and Limiting Information Exposure in DNN Layers
• Generalized Negative Correlation Learning for Deep Ensembling
• Building Deep, Equivariant Capsule Networks
• MMGAN: Generative Adversarial Networks for Multi-Modal Distributions
• Traces of Class/Cross-Class Structure Pervade Deep Learning Spectra
• Thief, Beware of What Get You There: Towards Understanding Model Extraction Attack
• Structured Bayesian Compression for Deep Neural Networks Based on The Turbo-VBI Approach
• Learning the Precise Feature for Cluster Assignment
• Clustering-Oriented Representation Learning with Attractive-Repulsive Loss
• Unsupervised Multi-Domain Image Translation with Domain-Specific Encoders/Decoders
• Analyzing Monotonic Linear Interpolation in Neural Network Loss Landscapes
• Evaluating State-of-the-Art Classification Models Against Bayes Optimality
• Better the Devil you Know: An Analysis of Evasion Attacks using Out-of-Distribution Adversarial Examples
• Second-Order Provable Defenses against Adversarial Attacks
• Training Federated GANs with Theoretical Guarantees: A Universal Aggregation Approach
• Be Like Water: Robustness to Extraneous Variables Via Adaptive Feature Normalization
• Constrained deep neural network architecture search for IoT devices accounting hardware calibration
• KerCNNs: biologically inspired lateral connections for classification of corrupted images
• Identifying Classes Susceptible to Adversarial Attacks
• Meta Feature Modulator for Long-tailed Recognition
• Semantic-Based Neural Network Repair
• Robust Training in High Dimensions via Block Coordinate Geometric Median Descent
• Bias-Free FedGAN: A Federated Approach to Generate Bias-Free Datasets
• Small Towers Make Big Differences
• Classifying the classifier: dissecting the weight space of neural networks
• Multi-VFL: A Vertical Federated Learning System for Multiple Data and Label Owners
• Multimodal Generative Models for Compositional Representation Learning
• NeuNetS: An Automated Synthesis Engine for Neural Network Design
• Adversarial Attack and Defense in Deep Ranking
• A Tale of Two Approximations: Tightening Over-Approximation for DNN Robustness Verification via Under-Approximation
• Inductive Transfer for Neural Architecture Optimization
• DeepOBS: A Deep Learning Optimizer Benchmark Suite
• The functional role of cue-driven feature-based feedback in object recognition
• Do We Really Need to Learn Representations from In-domain Data for Outlier Detection?
• Understanding the Effect of Bias in Deep Anomaly Detection
• AutoAssist: A Framework to Accelerate Training of Deep Neural Networks
• Interpretable Graph Capsule Networks for Object Recognition
• POPNASv2: An Efficient Multi-Objective Neural Architecture Search Technique
• Measuring Dependence with Matrix-based Entropy Functional
• Bayesian Inference Forgetting
• Approximating Instance-Dependent Noise via Instance-Confidence Embedding
• On the Effectiveness of Regularization Against Membership Inference Attacks
• Machine Learning Method for Functional Assessment of Retinal Models
• Consistent Representation Learning for High Dimensional Data Analysis
• A Quantum Edge Detection Algorithm
• Data Sanity Check for Deep Learning Systems via Learnt Assertions
• Unsupervised Data Imputation via Variational Inference of Deep Subspaces
• Generalized Zero-Shot Domain Adaptation via Coupled Conditional Variational Autoencoders
• QDCNN: Quantum Dilated Convolutional Neural Network
• Low-Latency Federated Learning over Wireless Channels with Differential Privacy
• Analytic Insights into Structure and Rank of Neural Network Hessian Maps
• Visual Transformer for Task-aware Active Learning
• Wasserstein-Wasserstein Auto-Encoders
• Leveraging Spatial and Temporal Correlations in Sparsified Mean Estimation
• Characteristics of Monte Carlo Dropout in Wide Neural Networks
• Learning from a Teacher using Unlabeled Data
• On Stein Variational Neural Network Ensembles
• Multi-class Generative Adversarial Nets for Semi-supervised Image Classification
• AdaSGD: Bridging the gap between SGD and Adam
• More Than Meets The Eye: Semi-supervised Learning Under Non-IID Data
• Clusterability in Neural Networks
• Using Dimensionality Reduction to Optimize t-SNE
• Learning Equivariant Energy Based Models with Equivariant Stein Variational Gradient Descent
• Is Shapley Value fair? Improving Client Selection for Mavericks in Federated Learning
• Quantitative Evaluation of Time-Dependent Multidimensional Projection Techniques
• Prediction is very hard, especially about conversion. Predicting user purchases from clickstream data in fashion e-commerce
• Deep Neural Architecture Search with Deep Graph Bayesian Optimization
• A Unified Plug-and-Play Framework for Effective Data Denoising and Robust Abstention
• Self-Supervised Representation Learning for Visual Anomaly Detection
• URSABench: Comprehensive Benchmarking of Approximate Bayesian Inference Methods for Deep Neural Networks
• Edge of chaos as a guiding principle for modern neural network training
• Training CNNs with Selective Allocation of Channels
• Topological Insights into Sparse Neural Networks
• Bayesian Tensor Network with Polynomial Complexity for Probabilistic Machine Learning
• A Generalizable Approach to Learning Optimizers
• Unbalanced Optimal Transport through Non-negative Penalized Linear Regression
• Epigenetic evolution of deep convolutional models
• MHVAE: a Human-Inspired Deep Hierarchical Generative Model for Multimodal Representation Learning
• Deep Contrastive Graph Representation via Adaptive Homotopy Learning
• SimEx: Express Prediction of Inter-dataset Similarity by a Fleet of Autoencoders
• Deep Active Learning with Augmentation-based Consistency Estimation
• Multi-Objective Neural Architecture Search Based on Diverse Structures and Adaptive Recommendation
• On Batch Normalisation for Approximate Bayesian Inference
• DIVINE: Diverse Influential Training Points for Data Visualization and Model Refinement
• Privacy-preserving Collaborative Learning with Automatic Transformation Search
• Label differential privacy via clustering
• Darks and Stripes: Effects of Clothing on Weight Perception
• Reconfigurable Intelligent Surface Assisted Mobile Edge Computing with Heterogeneous Learning Tasks
• Finding trainable sparse networks through Neural Tangent Transfer
• A Value-Function-based Interior-point Method for Non-convex Bi-level Optimization
• Learning Invariant Representation for Continual Learning
• Maximizing Information Gain in Partially Observable Environments via Prediction Reward
• Continual Learning with Fully Probabilistic Models
• The Impact of Reinitialization on Generalization in Convolutional Neural Networks
• Multitask Learning with Single Gradient Step Update for Task Balancing
• Overhead MNIST: A Benchmark Satellite Dataset
• Deep Generative Learning via Variational Gradient Flow
• Adversarial Robustness Guarantees for Gaussian Processes
• HufuNet: Embedding the Left Piece as Watermark and Keeping the Right Piece for Ownership Verification in Deep Neural Networks
• Global Convergence of Gradient Descent for Deep Linear Residual Networks
• Resource-constrained Federated Edge Learning with Heterogeneous Data: Formulation and Analysis
• The Connection between Out-of-Distribution Generalization and Privacy of ML Models
• k-GANs: Ensemble of Generative Models with Semi-Discrete Optimal Transport
• Training on test data: Removing near duplicates in Fashion-MNIST
• Active Generative Adversarial Network for Image Classification
• Reconstruction and Membership Inference Attacks against Generative Models
• Cross-modal Variational Auto-encoder with Distributed Latent Spaces and Associators
• Dueling Decoders: Regularizing Variational Autoencoder Latent Spaces
• Facilitating Bayesian Continual Learning by Natural Gradients and Stein Gradients
• Combining Model and Parameter Uncertainty in Bayesian Neural Networks
• Divergence Triangle for Joint Training of Generator Model, Energy-based Model, and Inference Model
• On the Learning Property of Logistic and Softmax Losses for Deep Neural Networks
• PixelHop++: A Small Successive-Subspace-Learning-Based (SSL-based) Model for Image Classification
• Intelligence, physics and information -- the tradeoff between accuracy and simplicity in machine learning
• White Noise Analysis of Neural Networks
• A Fully Stochastic Second-Order Trust Region Method
• An Effective Hit-or-Miss Layer Favoring Feature Interpretation as Learned Prototypes Deformations
• Perturbations are not Enough: Generating Adversarial Examples with Spatial Distortions
• PRI-VAE: Principle-of-Relevant-Information Variational Autoencoders
• Differentially Private Variational Autoencoders with Term-wise Gradient Aggregation
• Interpretable Visualizations with Differentiating Embedding Networks
• Efficient Computation Reduction in Bayesian Neural Networks Through Feature Decomposition and Memorization
• Learning Representations For Images With Hierarchical Labels
• A Review of Formal Methods applied to Machine Learning
• Exponentiated Gradient Reweighting for Robust Training Under Label Noise and Beyond
• Semi-supervised learning by selective training with pseudo labels via confidence estimation
• Measuring Discrimination to Boost Comparative Testing for Multiple Deep Learning Models
• PyTorch-Hebbian: facilitating local learning in a deep learning framework
• Removing Class Imbalance using Polarity-GAN: An Uncertainty Sampling Approach
• Advances in the training, pruning and enforcement of shape constraints of Morphological Neural Networks using Tropical Algebra
• Learning Translation Invariance in CNNs
• Dynamic Neural Diversification: Path to Computationally Sustainable Neural Networks
• Can convolutional ResNets approximately preserve input distances? A frequency analysis perspective
• Effective Evaluation of Deep Active Learning on Image Classification Tasks
• Regularization Can Help Mitigate Poisoning Attacks... with the Right Hyperparameters
• Time-multiplexed In-memory computation scheme for mapping Quantized Neural Networks on hybrid CMOS-OxRAM building blocks
• Simple, Scalable, and Stable Variational Deep Clustering
• Mapping the Potential and Pitfalls of "Data Dividends" as a Means of Sharing the Profits of Artificial Intelligence
• ExAD: An Ensemble Approach for Explanation-based Adversarial Detection
• Grouping Capsules Based Different Types
• Generalised Perceptron Learning
• Identifying Incorrect Classifications with Balanced Uncertainty
• A Probabilistic Representation of Deep Learning for Improving The Information Theoretic Interpretability
• Blind Adversarial Training: Balance Accuracy and Robustness
• Measuring the Complexity of Domains Used to Evaluate AI Systems
• Performance evaluation of an integrated photonic convolutional neural network based on delay buffering and wavelength division multiplexing
• Ask-n-Learn: Active Learning via Reliable Gradient Representations for Image Classification
• MIM: Mutual Information Machine
• Learning to Remember from a Multi-Task Teacher
• Unsupervised Domain Adaptation using Deep Networks with Cross-Grafted Stacks
• Energy-Inspired Models: Learning with Sampler-Induced Distributions
• Incremental Concept Learning via Online Generative Memory Recall
• Gaussian Process Bandit Optimization of the Thermodynamic Variational Objective
• On Masked Pre-training and the Marginal Likelihood
• Pre-interpolation loss behaviour in neural networks
• DeepSwarm: Optimising Convolutional Neural Networks using Swarm Intelligence
• On the Evolution of Neuron Communities in a Deep Learning Architecture
• Routing Towards Discriminative Power of Class Capsules
• Why is the prediction wrong? Towards underfitting case explanation via meta-classification
• Rectangular Flows for Manifold Learning
• Optimization Variance: Exploring Generalization Properties of DNNs
• Learning Generative Models of Structured Signals from Their Superposition Using GANs with Application to Denoising and Demixing
• Robust Semantic Interpretability: Revisiting Concept Activation Vectors
• Enabling Inference Privacy with Adaptive Noise Injection
• Asynchronous Distributed Optimization with Redundancy in Cost Functions
• Probabilistic Discriminative Learning with Layered Graphical Models
• Bespoke vs. Prêt-à-Porter Lottery Tickets: Exploiting Mask Similarity for Trainable Sub-Network Finding
• A Probabilistic Representation of DNNs: Bridging Mutual Information and Generalization
• DAC: Deep Autoencoder-based Clustering, a General Deep Learning Framework of Representation Learning
• Overhead-MNIST: Machine Learning Baselines for Image Classification
• On Out-of-distribution Detection with Energy-based Models
• Semi-Supervised Class Discovery
• A Multiclass Classification Approach to Label Ranking
• On Positive-Unlabeled Classification in GAN
• Shapley Explanation Networks
• A Computationally Efficient Method for Defending Adversarial Deep Learning Attacks
• Improving the Evaluation of Generative Models with Fuzzy Logic
• Model-agnostic out-of-distribution detection using combined statistical tests
• TiFL: A Tier-based Federated Learning System
• Convolutional Dictionary Pair Learning Network for Image Representation Learning
• Transfer Learning in General Lensless Imaging through Scattering Media
• Gravity Optimizer: a Kinematic Approach on Optimization in Deep Learning
• How many winning tickets are there in one DNN?
• Generative Adversarial Data Programming
• Interpreting the Latent Space of GANs via Correlation Analysis for Controllable Concept Manipulation
• Selective Forgetting of Deep Networks at a Finer Level than Samples
• Iterative Distillation for Better Uncertainty Estimates in Multitask Emotion Recognition
• Tutorials on Testing Neural Networks
• DarkGAN: Exploiting Knowledge Distillation for Comprehensible Audio Synthesis with GANs
• Investigating a Baseline Of Self Supervised Learning Towards Reducing Labeling Costs For Image Classification
• ImagiFilter: A resource to enable the semi-automatic mining of images at scale
• Single Sample Feature Importance: An Interpretable Algorithm for Low-Level Feature Analysis
• PAC-Bayesian Generalization Bounds for MultiLayer Perceptrons
• On Focal Loss for Class-Posterior Probability Estimation: A Theoretical Perspective
• Prior Activation Distribution (PAD): A Versatile Representation to Utilize DNN Hidden Units
• On the Robustness of Active Learning
• On the One-sided Convergence of Adam-type Algorithms in Non-convex Non-concave Min-max Optimization
• Self-Evolutionary Optimization for Pareto Front Learning
• Channel-Directed Gradients for Optimization of Convolutional Neural Networks
• Tensor Network for Supervised Learning at Finite Temperature
• Ordering Dimensions with Nested Dropout Normalizing Flows
• Effective Version Space Reduction for Convolutional Neural Networks
• Generating Unrestricted Adversarial Examples via Three Parameters
• Experiments with Rich Regime Training for Deep Learning
• Generating Adversarial Examples with an Optimized Quality
• Efficient Proximal Mapping of the 1-path-norm of Shallow Networks
• Domain2Vec: Domain Embedding for Unsupervised Domain Adaptation
• SPENSER: Towards a NeuroEvolutionary Approach for Convolutional Spiking Neural Networks
• FedNS: Improving Federated Learning for collaborative image classification on mobile clients
• Adma: A Flexible Loss Function for Neural Networks
• On the Intrinsic Differential Privacy of Bagging
• iCVI-ARTMAP: Accelerating and improving clustering using adaptive resonance theory predictive mapping and incremental cluster validity indices
• ALEX: Active Learning based Enhancement of a Model's Explainability
• Adversarial Learning for Zero-shot Domain Adaptation
• A Study on the Uncertainty of Convolutional Layers in Deep Neural Networks
• NeVer 2.0: Learning, Verification and Repair of Deep Neural Networks
• LADA: Look-Ahead Data Acquisition via Augmentation for Active Learning
• Efficient Multimodal Sampling via Tempered Distribution Flow
• Towards Interaction Detection Using Topological Analysis on Neural Networks
• Bilevel Continual Learning
• Self-paced Data Augmentation for Training Neural Networks
• Teaching a GAN What Not to Learn
• Optimal Transport Based Generative Autoencoders
• CARMS: Categorical-Antithetic-REINFORCE Multi-Sample Gradient Estimator
• Mapper Based Classifier
• Instance-Dependent Partial Label Learning
• Gradient-based Quadratic Multiform Separation
• Signature-Graph Networks
• Disturbing Target Values for Neural Network Regularization
• Ghost Loss to Question the Reliability of Training Data
• Multi-level Similarity Learning for Low-Shot Recognition
• Angular Learning: Toward Discriminative Embedded Features
• A fast asynchronous MCMC sampler for sparse Bayesian inference
• DeGAN : Data-Enriching GAN for Retrieving Representative Samples from a Trained Classifier
• Reconstructing Images of Two Adjacent Objects through Scattering Medium Using Generative Adversarial Network
• eProduct: A Million-Scale Visual Search Benchmark to Address Product Recognition Challenges
• Towards Robust Active Feature Acquisition
• Effects of VLSI Circuit Constraints on Temporal-Coding Multilayer Spiking Neural Networks
• HyperNP: Interactive Visual Exploration of Multidimensional Projection Hyperparameters
• PatentNet: A Large-Scale Incomplete Multiview, Multimodal, Multilabel Industrial Goods Image Database
• Mapper Comparison with Wasserstein Metrics
• Orthogonal-Padé Activation Functions: Trainable Activation functions for smooth and faster convergence in deep networks
• Robust Out-of-Distribution Detection on Deep Probabilistic Generative Models
• Deep Supervised Hashing leveraging Quadratic Spherical Mutual Information for Content-based Image Retrieval
• MONCAE: Multi-Objective Neuroevolution of Convolutional Autoencoders
• Venn GAN: Discovering Commonalities and Particularities of Multiple Distributions
• On Transformations in Stochastic Gradient MCMC
• DPER: Efficient Parameter Estimation for Randomly Missing Data
• Top-$k$ Regularization for Supervised Feature Selection
• H-FL: A Hierarchical Communication-Efficient and Privacy-Protected Architecture for Federated Learning
• On the Exploitation of Neuroevolutionary Information: Analyzing the Past for a More Efficient Future
• Learning Generative Prior with Latent Space Sparsity Constraints
• Novel Uncertainty Framework for Deep Learning Ensembles
• Exploring layerwise decision making in DNNs
• Revisiting hard thresholding for DNN pruning
• Learning to Optimize with Dynamic Mode Decomposition
• Context-modulation of hippocampal dynamics and deep convolutional networks
• Hierarchical Annotation of Images with Two-Alternative-Forced-Choice Metric Learning
• Learning in Deep Neural Networks Using a Biologically Inspired Optimizer
• Stochastic Bottleneck: Rateless Auto-Encoder for Flexible Dimensionality Reduction
• Noise Homogenization via Multi-Channel Wavelet Filtering for High-Fidelity Sample Generation in GANs
• Bottom-up Hierarchical Classification Using Confusion-based Logit Compression
• Feedback Control for Online Training of Neural Networks
• Soft Compression for Lossless Image Coding
• Multi-class Probabilistic Bounds for Self-learning
• DOODLER: Determining Out-Of-Distribution Likelihood from Encoder Reconstructions
• A novel network training approach for open set image recognition
• Contributions to Large Scale Bayesian Inference and Adversarial Machine Learning
• Open-World Class Discovery with Kernel Networks
• SimpleBooks: Long-term dependency book dataset with simplified English vocabulary for word-level language modeling
• Mitigating Sampling Bias and Improving Robustness in Active Learning
• Investigating the Effect of Intraclass Variability in Temporal Ensembling
• Task Guided Compositional Representation Learning for ZDA
• Magnitude and Uncertainty Pruning Criterion for Neural Networks
• Automated Robustness with Adversarial Training as a Post-Processing Step
• Generative Adversarial Networks with Inverse Transformation Unit
• Using Topological Framework for the Design of Activation Function and Model Pruning in Deep Neural Networks
• Simple and Scalable Parallelized Bayesian Optimization
• CrossedWires: A Dataset of Syntactically Equivalent but Semantically Disparate Deep Learning Models
• ICLR 2021 Challenge for Computational Geometry & Topology: Design and Results
• Out-of-Distribution Example Detection in Deep Neural Networks using Distance to Modelled Embedding
• Towards More Efficient Federated Learning with Better Optimization Objects
• Incremental Verification of Fixed-Point Implementations of Neural Networks
• Shared and Private VAEs with Generative Replay for Continual Learning
• Efficient Byzantine-Resilient Stochastic Gradient Desce
• Part-based approximations for morphological operators using asymmetric auto-encoders
• Practical Insights of Repairing Model Problems on Image Classification
• On Addressing Heterogeneity in Federated Learning for Autonomous Vehicles Connected to a Drone Orchestrator
• Cross-Domain Latent Modulation for Variational Transfer Learning
• Variance Regularization for Accelerating Stochastic Optimization
• From Boltzmann Machines to Neural Networks and Back Again
• MTL2L: A Context Aware Neural Optimiser
• Explicit Gradient Learning
• Visualizing the Finer Cluster Structure of Large-Scale and High-Dimensional Data
• Leave-one-out Unfairness
• DNNs as Layers of Cooperating Classifiers
• Inertial Proximal Deep Learning Alternating Minimization for Efficient Neutral Network Training
• Towards Sharper First-Order Adversary with Quantized Gradients
• Condensed Composite Memory Continual Learning
• ALT-MAS: A Data-Efficient Framework for Active Testing of Machine Learning Algorithms
• Large-data determinantal clustering
• A Non-Intrusive Correction Algorithm for Classification Problems with Corrupted Data
• Nested Learning For Multi-Granular Tasks
• Nonlinear Least Squares for Large-Scale Machine Learning using Stochastic Jacobian Estimates
• Label Smoothed Embedding Hypothesis for Out-of-Distribution Detection
• Lifelong Mixture of Variational Autoencoders
• Learning to Utilize Correlated Auxiliary Noise: A Possible Quantum Advantage
• Cluster Representatives Selection in Non-Metric Spaces for Nearest Prototype Classification
• Learning Gaussian-Bernoulli RBMs using Difference of Convex Functions Optimization
• Towards Robust Classification with Deep Generative Forests
• On the generalization of bayesian deep nets for multi-class classification
• How many samples are needed to reliably approximate the best linear estimator for a linear inverse problem?
• Multi-objective Evolutionary Approach for Efficient Kernel Size and Shape for CNN
• On Intrinsic Dataset Properties for Adversarial Machine Learning
• Random Graph-Based Neuromorphic Learning with a Layer-Weaken Structure
• Coded-InvNet for Resilient Prediction Serving Systems
• Efficient Tensor Contraction via Fast Count Sketch
• Target Propagation via Regularized Inversion
• Learning Under Delayed Feedback: Implicitly Adapting to Gradient Delays
• Network-Density-Controlled Decentralized Parallel Stochastic Gradient Descent in Wireless Systems
• Improve SGD Training via Aligning Mini-batches
• Rethinking Layer-wise Feature Amounts in Convolutional Neural Network Architectures
• Novelty Detection via Contrastive Learning with Negative Data Augmentation
• Auto-CASH: Autonomous Classification Algorithm Selection with Deep Q-Network
• DaiMoN: A Decentralized Artificial Intelligence Model Network
• Is the U-Net Directional-Relationship Aware?
• Stochastic Doubly Robust Gradient
• Stochastic gradient algorithms from ODE splitting perspective
• Interpretable BoW Networks for Adversarial Example Detection
• Robust Unsupervised Multi-Object Tracking in Noisy Environments
• Probabilistic Decoupling of Labels in Classification
• Non-Gradient Manifold Neural Network
• A Trainable Multiplication Layer for Auto-correlation and Co-occurrence Extraction
• Atlas Based Representation and Metric Learning on Manifolds
• Virtual Conditional Generative Adversarial Networks
• An Ensemble Approach Towards Adversarial Robustness
• Group Equivariant Subsampling
• Local Convergence of Adaptive Gradient Descent Optimizers
• Widely Linear Kernels for Complex-Valued Kernel Activation Functions
• RSAC: Regularized Subspace Approximation Classifier for Lightweight Continuous Learning
• Stochastic Computing for Hardware Implementation of Binarized Neural Networks
• Consensus Driven Learning
• Tomographic Auto-Encoder: Unsupervised Bayesian Recovery of Corrupted Data
• Efficient Data-Dependent Learnability
• Spatially Correlated Patterns in Adversarial Images
• Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks
• Prior-Independent Auctions for the Demand Side of Federated Learning
• Learning a Deep Generative Model like a Program: the Free Category Prior
• CatFedAvg: Optimising Communication-efficiency and Classification Accuracy in Federated Learning
• Understanding Classifier Mistakes with Generative Models
• Learning Spatial Relationships between Samples of Patent Image Shapes
• Detecting Deep Neural Network Defects with Data Flow Analysis
• Stochastic Sparse Learning with Momentum Adaptation for Imprecise Memristor Networks
• Conditioned Text Generation with Transfer for Closed-Domain Dialogue Systems
• Self-Checking Deep Neural Networks in Deployment
• A $0.11-0.38$ pJ/cycle Differential Ring Oscillator in $65$ nm CMOS for Robust Neurocomputing
• Trust but Verify: Assigning Prediction Credibility by Counterfactual Constrained Learning
• Learning from Incomplete Features by Simultaneous Training of Neural Networks and Sparse Coding
• PipeTune: Pipeline Parallelism of Hyper and System Parameters Tuning for Deep Learning Clusters
• Facilitate the Parametric Dimension Reduction by Gradient Clipping
• Loosely Coupled Federated Learning Over Generative Models
• Regularizing Neural Networks via Stochastic Branch Layers
• Over-parametrized neural networks as under-determined linear systems
• Vanishing Twin GAN: How training a weak Generative Adversarial Network can improve semi-supervised image classification
• The Effect of Optimization Methods on the Robustness of Out-of-Distribution Detection Approaches
• About contrastive unsupervised representation learning for classification and its convergence
• High Mutual Information in Representation Learning with Symmetric Variational Inference
• Quantum Algorithms for Unsupervised Machine Learning and Neural Networks
• Adversarial Training: embedding adversarial perturbations into the parameter space of a neural network to build a robust system
• Optimizing Information-theoretical Generalization Bounds via Anisotropic Noise in SGLD
• CheckNet: Secure Inference on Untrusted Devices
• How to boost autoencoders?
• neuralRank: Searching and ranking ANN-based model repositories
• Exact Stochastic Second Order Deep Learning
• Generating Adversarial Examples With Conditional Generative Adversarial Net
• Targeted Deep Learning: Framework, Methods, and Applications
• A Closer Look at Reference Learning for Fourier Phase Retrieval
• Lifelong Learning Process: Self-Memory Supervising and Dynamically Growing Networks
• Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learning
• Dendritic Self-Organizing Maps for Continual Learning
• Computational Analysis of Deformable Manifolds: from Geometric Modelling to Deep Learning
• Pixel-wise Conditioning of Generative Adversarial Networks
• Spatially-Coupled Neural Network Architectures
• Does the Layout Really Matter? A Study on Visual Model Accuracy Estimation
• The Neglected Sibling: Isotropic Gaussian Posterior for VAE
• Disassembling Object Representations without Labels
• Deep Unsupervised Image Anomaly Detection: An Information Theoretic Framework
• Likelihood Landscapes: A Unifying Principle Behind Many Adversarial Defenses
• Generalizing Neural Networks by Reflecting Deviating Data in Production