Improved Training of Wasserstein GANs

Abstract

Generative Adversarial Networks (GANs) are powerful generative models, but suffer from training instability. The recently proposed Wasserstein GAN (WGAN) makes progress toward stable training of GANs, but sometimes can still generate only low-quality samples or fail to converge. We find that these problems are often due to the use of weight clipping in WGAN to enforce a Lipschitz constraint on the critic, which can lead to undesired behavior. We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input. Our proposed method performs better than standard WGAN and enables stable training of a wide variety of GAN architectures with almost no hyperparameter tuning, including 101-layer ResNets and language models over discrete data. We also achieve high quality generations on CIFAR-10 and LSUN bedrooms.

arXiv commentNIPS camera-ready

References in corpus (23)

• Adam: A Method for Stochastic Optimization
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Categorical Reparameterization with Gumbel-Softmax
• Instance Normalization: The Missing Ingredient for Fast Stylization
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Conditional Image Generation with PixelCNN Decoders
• Improved Techniques for Training GANs
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• BEGAN: Boundary Equilibrium Generative Adversarial Networks
• Adversarially Learned Inference
• f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization
• Unrolled Generative Adversarial Networks
• SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient
• Adversarial Learning for Neural Dialogue Generation
• Maximum-Likelihood Augmented Discrete Generative Adversarial Networks
• Least Squares Generative Adversarial Networks
• Boundary-Seeking Generative Adversarial Networks
• On distinguishability criteria for estimating generative models
• Learning to Draw Samples: With Application to Amortized MLE for Generative Adversarial Learning
• Recurrent Topic-Transition GAN for Visual Paragraph Generation
• Improved generator objectives for GANs
• Stacked Generative Adversarial Networks

Cited by in corpus (462)

• Spectral Normalization for Generative Adversarial Networks
• Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
• Recent advances and clinical applications of deep learning in medical image analysis
• A Generative Model for Inverse Design of Metamaterials
• Training Generative Adversarial Networks with Limited Data
• Data Augmentation Generative Adversarial Networks
• The relativistic discriminator: a key element missing from standard GAN
• Deep Reinforcement Learning: An Overview
• MolGAN: An implicit generative model for small molecular graphs
• Generating Natural Adversarial Examples
• Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?
• MMD GAN: Towards Deeper Understanding of Moment Matching Network
• cGANs with Projection Discriminator
• InfoVAE: Information Maximizing Variational Autoencoders
• Sharpness-aware Low dose CT denoising using conditional generative adversarial network
• Differentially Private Generative Adversarial Network
• Generating 3D structures from a 2D slice with GAN-based dimensionality expansion
• Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Generative Adversarial Network for Medical Images (MI-GAN)
• The Cramer Distance as a Solution to Biased Wasserstein Gradients
• Image Inpainting via Generative Multi-column Convolutional Neural Networks
• Variational Approaches for Auto-Encoding Generative Adversarial Networks
• Generating Adversarial Examples with Adversarial Networks
• An empirical study on evaluation metrics of generative adversarial networks
• Learning to Predict the Cosmological Structure Formation
• Generative Image Inpainting with Contextual Attention
• Synthesizing Tabular Data using Generative Adversarial Networks
• Stabilizing Training of Generative Adversarial Networks through Regularization
• A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecasts
• Continual Learning with Deep Generative Replay
• NetGAN: Generating Graphs via Random Walks
• Wasserstein Auto-Encoders
• The Numerics of GANs
• How to GAN LHC Events
• Masking: A New Perspective of Noisy Supervision
• An Introduction to Image Synthesis with Generative Adversarial Nets
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• TomoGAN: Low-Dose Synchrotron X-Ray Tomography with Generative Adversarial Networks
• Deep Identity-aware Transfer of Facial Attributes
• Selective Synthetic Augmentation with HistoGAN for Improved Histopathology Image Classification
• Towards the Automatic Anime Characters Creation with Generative Adversarial Networks
• On the regularization of Wasserstein GANs
• A survey on text generation using generative adversarial networks
• Generative Adversarial Active Learning
• SPG-Net: Segmentation Prediction and Guidance Network for Image Inpainting
• Video-to-Video Synthesis
• Unsupervised Detection of Lesions in Brain MRI using constrained adversarial auto-encoders
• Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GAN
• Unsupervised Learning for Cell-level Visual Representation in Histopathology Images with Generative Adversarial Networks
• Semi-supervised Learning with GANs: Manifold Invariance with Improved Inference
• Visual enhancement of Cone-beam CT by use of CycleGAN
• Boundary-Seeking Generative Adversarial Networks
• From Variational to Deterministic Autoencoders
• AI-assisted super-resolution cosmological simulations
• Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Step
• Adversarial Generation of Natural Language
• Generative Models for Automatic Chemical Design
• Fast and accurate simulation of particle detectors using generative adversarial networks
• Language Generation with Recurrent Generative Adversarial Networks without Pre-training
• Beyond Self-attention: External Attention using Two Linear Layers for Visual Tasks
• Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
• ChatPainter: Improving Text to Image Generation using Dialogue
• On Accurate Evaluation of GANs for Language Generation
• Adversarial Deep Averaging Networks for Cross-Lingual Sentiment Classification
• A-NICE-MC: Adversarial Training for MCMC
• Learning Generative Models with Sinkhorn Divergences
• Pedestrian-Synthesis-GAN: Generating Pedestrian Data in Real Scene and Beyond
• Super-resolution emulator of cosmological simulations using deep physical models
• On the Effects of Batch and Weight Normalization in Generative Adversarial Networks
• Geometry Score: A Method For Comparing Generative Adversarial Networks
• Stabilizing Generative Adversarial Networks: A Survey
• Improved Adversarial Systems for 3D Object Generation and Reconstruction
• Recent Developments Combining Ensemble Smoother and Deep Generative Networks for Facies History Matching
• Feedback GAN (FBGAN) for DNA: a Novel Feedback-Loop Architecture for Optimizing Protein Functions
• Wasserstein Learning of Deep Generative Point Process Models
• Detecting outliers in astronomical images with deep generative networks
• Sliced-Wasserstein Autoencoder: An Embarrassingly Simple Generative Model
• TBD: Benchmarking and Analyzing Deep Neural Network Training
• MAGAN: Margin Adaptation for Generative Adversarial Networks
• Pseudo-healthy synthesis with pathology disentanglement and adversarial learning
• On gradient regularizers for MMD GANs
• Exposure: A White-Box Photo Post-Processing Framework
• Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect
• Learning Independent Features with Adversarial Nets for Non-linear ICA
• Geometry-Contrastive GAN for Facial Expression Transfer
• Is Generator Conditioning Causally Related to GAN Performance?
• Activation Maximization Generative Adversarial Nets
• Convergence Problems with Generative Adversarial Networks (GANs)
• Generating Multi-Categorical Samples with Generative Adversarial Networks
• Conditional Sig-Wasserstein GANs for Time Series Generation
• TWIST-GAN: Towards Wavelet Transform and Transferred GAN for Spatio-Temporal Single Image Super Resolution
• Comparison of Maximum Likelihood and GAN-based training of Real NVPs
• Void Filling of Digital Elevation Models with Deep Generative Models
• Generative Models for Low-Rank Video Representation and Reconstruction
• CR-GAN: Learning Complete Representations for Multi-view Generation
• Probability-Density-Based Deep Learning Paradigm for the Fuzzy Design of Functional Metastructures
• AGAN: Towards Automated Design of Generative Adversarial Networks
• ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
• Adversarially Regularized Autoencoders
• A Convex Duality Framework for GANs
• SC-FEGAN: Face Editing Generative Adversarial Network with User's Sketch and Color
• Generative Adversarial Network Training is a Continual Learning Problem
• Text2Shape: Generating Shapes from Natural Language by Learning Joint Embeddings
• SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis
• Anomaly detection and automatic labeling for solar cell quality inspection based on Generative Adversarial Network
• PID-GAN: A GAN Framework based on a Physics-informed Discriminator for Uncertainty Quantification with Physics
• Understanding GANs: the LQG Setting
• HIGAN: Cosmic Neutral Hydrogen with Generative Adversarial Networks
• Improving Detection of Credit Card Fraudulent Transactions using Generative Adversarial Networks
• Decentralized and Robust Privacy-Preserving Model Using Blockchain-Enabled Federated Deep Learning in Intelligent Enterprises
• Convolutional Generative Adversarial Networks with Binary Neurons for Polyphonic Music Generation
• Design by adaptive sampling
• Class-Splitting Generative Adversarial Networks
• Text to Image Synthesis Using Generative Adversarial Networks
• Airline Passenger Name Record Generation using Generative Adversarial Networks
• Stabilizing Invertible Neural Networks Using Mixture Models
• Adversarial Distillation of Bayesian Neural Network Posteriors
• Stochastic Restoration of Heavily Compressed Musical Audio using Generative Adversarial Networks
• CA-GAN: Weakly Supervised Color Aware GAN for Controllable Makeup Transfer
• AutoLoss: Learning Discrete Schedules for Alternate Optimization
• Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields
• Masked Language Modeling for Proteins via Linearly Scalable Long-Context Transformers
• Deep Encoder-decoder Adversarial Reconstruction (DEAR) Network for 3D CT from Few-view Data
• On the Benefit of Adversarial Training for Monocular Depth Estimation
• Improved Training of Generative Adversarial Networks Using Representative Features
• Perceptual Image Anomaly Detection
• From EMBER to FIRE: predicting high resolution baryon fields from dark matter simulations with Deep Learning
• 3D Object Reconstruction from a Single Depth View with Adversarial Learning
• Low-Resource Text Classification using Domain-Adversarial Learning
• Memorization Precedes Generation: Learning Unsupervised GANs with Memory Networks
• GAN-QP: A Novel GAN Framework without Gradient Vanishing and Lipschitz Constraint
• DeepFlow: History Matching in the Space of Deep Generative Models
• Synthesizing realistic neural population activity patterns using Generative Adversarial Networks
• Structured Generative Adversarial Networks
• Image to Image Translation for Domain Adaptation
• Continuous conditional generative adversarial networks for data-driven solutions of poroelasticity with heterogeneous material properties
• A Divergence Minimization Perspective on Imitation Learning Methods
• Unsupervised Cipher Cracking Using Discrete GANs
• MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks
• Generative Adversarial Networks: A Survey Towards Private and Secure Applications
• Generative Adversarial Network Architectures For Image Synthesis Using Capsule Networks
• Neural Language Generation: Formulation, Methods, and Evaluation
• Cleaning our own Dust: Simulating and Separating Galactic Dust Foregrounds with Neural Networks
• Mode Collapse and Regularity of Optimal Transportation Maps
• Attacks on State-of-the-Art Face Recognition using Attentional Adversarial Attack Generative Network
• Adversarial Text Generation Without Reinforcement Learning
• Image Clustering using an Augmented Generative Adversarial Network and Information Maximization
• A unified framework for 21cm tomography sample generation and parameter inference with Progressively Growing GANs
• Decompose to manipulate: Manipulable Object Synthesis in 3D Medical Images with Structured Image Decomposition
• 2-Wasserstein Approximation via Restricted Convex Potentials with Application to Improved Training for GANs
• Stability-certified reinforcement learning: A control-theoretic perspective
• Self-Supervised Variational Auto-Encoders
• Attacking Speaker Recognition With Deep Generative Models
• Disentangled Makeup Transfer with Generative Adversarial Network
• Random Matrix Theory Proves that Deep Learning Representations of GAN-data Behave as Gaussian Mixtures
• Latent Convolutional Models
• Adversarial Variational Optimization of Non-Differentiable Simulators
• Towards a Better Understanding and Regularization of GAN Training Dynamics
• Exploring the Evolution of GANs through Quality Diversity
• eCommerceGAN : A Generative Adversarial Network for E-commerce
• Enhancing Mechanical Metamodels with a Generative Model-Based Augmented Training Dataset
• GANGs: Generative Adversarial Network Games
• What Matters for Adversarial Imitation Learning?
• A Generative Model for Texture Synthesis based on Optimal Transport between Feature Distributions
• Adversarial Contrastive Estimation
• Metropolis-Hastings Generative Adversarial Networks
• MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition
• An error analysis of generative adversarial networks for learning distributions
• O-GAN: Extremely Concise Approach for Auto-Encoding Generative Adversarial Networks
• On the Effectiveness of Least Squares Generative Adversarial Networks
• Compressing PDF sets using generative adversarial networks
• Robust learning with implicit residual networks
• Learned convex regularizers for inverse problems
• Finding Mixed Nash Equilibria of Generative Adversarial Networks
• Generative Adversarial Networks (GANs): What it can generate and What it cannot?
• Twin-GAN -- Unpaired Cross-Domain Image Translation with Weight-Sharing GANs
• SALSA-TEXT : self attentive latent space based adversarial text generation
• Conditioning of three-dimensional generative adversarial networks for pore and reservoir-scale models
• Deep Generative Models for Distribution-Preserving Lossy Compression
• DOOM Level Generation using Generative Adversarial Networks
• Packet-Level Adversarial Network Traffic Crafting using Sequence Generative Adversarial Networks
• Rethinking Sampling in 3D Point Cloud Generative Adversarial Networks
• Triple consistency loss for pairing distributions in GAN-based face synthesis
• Common Spatial Generative Adversarial Networks based EEG Data Augmentation for Cross-Subject Brain-Computer Interface
• Local and non-local dependency learning and emergence of rule-like representations in speech data by Deep Convolutional Generative Adversarial Networks
• Dual Swap Disentangling
• CMB-GAN: Fast Simulations of Cosmic Microwave background anisotropy maps using Deep Learning
• Accelerating Bayesian microseismic event location with deep learning
• RIGA: Covert and Robust White-Box Watermarking of Deep Neural Networks
• Quantitatively Evaluating GANs With Divergences Proposed for Training
• Deep Learning of Dynamic Subsurface Flow via Theory-guided Generative Adversarial Network
• Blood Vessel Geometry Synthesis using Generative Adversarial Networks
• Disentangling Multiple Conditional Inputs in GANs
• Improving the quality of generative models through Smirnov transformation
• Towards Unsupervised Automatic Speech Recognition Trained by Unaligned Speech and Text only
• Discriminative Region Proposal Adversarial Networks for High-Quality Image-to-Image Translation
• LOGAN: Membership Inference Attacks Against Generative Models
• TequilaGAN: How to easily identify GAN samples
• Semi-Supervised Multitask Learning on Multispectral Satellite Images Using Wasserstein Generative Adversarial Networks (GANs) for Predicting Poverty
• Learning Shape Priors for Single-View 3D Completion and Reconstruction
• Densely connected normalizing flows
• RadialGAN: Leveraging multiple datasets to improve target-specific predictive models using Generative Adversarial Networks
• Composite Functional Gradient Learning of Generative Adversarial Models
• Modular Generative Adversarial Networks
• New Era of Deeplearning-Based Malware Intrusion Detection: The Malware Detection and Prediction Based On Deep Learning
• GAN-powered Deep Distributional Reinforcement Learning for Resource Management in Network Slicing
• Bounded Information Rate Variational Autoencoders
• Primal-Dual Wasserstein GAN
• Unsupervised Anomaly Detection with Adversarial Mirrored AutoEncoders
• Fluid Flow Mass Transport for Generative Networks
• Adversarial Image Translation: Unrestricted Adversarial Examples in Face Recognition Systems
• Learning Deep Hidden Nonlinear Dynamics from Aggregate Data
• Generative Adversarial Imitation Learning for End-to-End Autonomous Driving on Urban Environments
• Improved Training with Curriculum GANs
• Training Generative Adversarial Networks Via Turing Test
• Subgrid-scale parametrization of unresolved scales in forced Burgers equation using Generative Adversarial Networks (GAN)
• Non-parametric estimation of Jensen-Shannon Divergence in Generative Adversarial Network training
• Optimal transport maps for distribution preserving operations on latent spaces of Generative Models
• Comparing Generative Adversarial Network Techniques for Image Creation and Modification
• Using generative adversarial networks to synthesize artificial financial datasets
• Improving LIME Robustness with Smarter Locality Sampling
• 3D-PhysNet: Learning the Intuitive Physics of Non-Rigid Object Deformations
• Cycle In Cycle Generative Adversarial Networks for Keypoint-Guided Image Generation
• Reconstruction of Protein Structures from Single-Molecule Time Series
• Data-driven Seismic Waveform Inversion: A Study on the Robustness and Generalization
• Towards Diverse Paraphrase Generation Using Multi-Class Wasserstein GAN
• Multi-target Voice Conversion without Parallel Data by Adversarially Learning Disentangled Audio Representations
• The Devil is in the Decoder: Classification, Regression and GANs
• General-to-Detailed GAN for Infrequent Class Medical Images
• Training Generative Adversarial Networks via Primal-Dual Subgradient Methods: A Lagrangian Perspective on GAN
• Generative Adversarial Networks for Crystal Structure Prediction
• Face Translation between Images and Videos using Identity-aware CycleGAN
• Deriving Machine Attention from Human Rationales
• Learning Style-Aware Symbolic Music Representations by Adversarial Autoencoders
• Wasserstein Variational Inference
• ACtuAL: Actor-Critic Under Adversarial Learning
• Improving GAN Training via Binarized Representation Entropy (BRE) Regularization
• Generating Synthetic Multispectral Satellite Imagery from Sentinel-2
• Understanding the Effectiveness of Lipschitz-Continuity in Generative Adversarial Nets
• RSM-GAN: A Convolutional Recurrent GAN for Anomaly Detection in Contaminated Seasonal Multivariate Time Series
• Few-shot Video-to-Video Synthesis
• Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning
• QSMGAN: Improved Quantitative Susceptibility Mapping using 3D Generative Adversarial Networks with Increased Receptive Field
• Recent Advances of Image Steganography with Generative Adversarial Networks
• Improving Conditional Sequence Generative Adversarial Networks by Stepwise Evaluation
• KG-GAN: Knowledge-Guided Generative Adversarial Networks
• Self-Supervised GANs with Label Augmentation
• DeepMasterPrints: Generating MasterPrints for Dictionary Attacks via Latent Variable Evolution
• Causal-TGAN: Generating Tabular Data Using Causal Generative Adversarial Networks
• Improving Movement Predictions of Traffic Actors in Bird's-Eye View Models using GANs and Differentiable Trajectory Rasterization
• Learning the Precise Feature for Cluster Assignment
• Normalized Wasserstein Distance for Mixture Distributions with Applications in Adversarial Learning and Domain Adaptation
• Discriminator Feature-based Inference by Recycling the Discriminator of GANs
• Image-derived generative modeling of pseudo-macromolecular structures - towards the statistical assessment of Electron CryoTomography template matching
• GANs beyond divergence minimization
• Uncertainty quantification in a mechanical submodel driven by a Wasserstein-GAN
• Inverse Estimation of Elastic Modulus Using Physics-Informed Generative Adversarial Networks
• Shaping Rewards for Reinforcement Learning with Imperfect Demonstrations using Generative Models
• PAGAN: Portfolio Analysis with Generative Adversarial Networks
• Statistical analysis of Wasserstein GANs with applications to time series forecasting
• Imitating Targets from all sides: An Unsupervised Transfer Learning method for Person Re-identification
• Performing Co-Membership Attacks Against Deep Generative Models
• Metric Learning-based Generative Adversarial Network
• Anomaly detection with Wasserstein GAN
• Neural Synthesis of Footsteps Sound Effects with Generative Adversarial Networks
• GAN-based Generation and Automatic Selection of Explanations for Neural Networks
• Learning to Synthesize Fashion Textures
• Gaussian-Smooth Optimal Transport: Metric Structure and Statistical Efficiency
• Variance Regularizing Adversarial Learning
• WGANVO: Monocular Visual Odometry based on Generative Adversarial Networks
• Controllable Descendant Face Synthesis
• BiHMP-GAN: Bidirectional 3D Human Motion Prediction GAN
• A Wasserstein GAN model with the total variational regularization
• toon2real: Translating Cartoon Images to Realistic Images
• SeismoGen: Seismic Waveform Synthesis Using Generative Adversarial Networks
• Unsupervised 3D Shape Learning from Image Collections in the Wild
• Unsupervised Change Detection in Satellite Images with Generative Adversarial Network
• Exploiting Local Feature Patterns for Unsupervised Domain Adaptation
• Re-using Adversarial Mask Discriminators for Test-time Training under Distribution Shifts
• AMIL: Adversarial Multi Instance Learning for Human Pose Estimation
• Towards Better Long-range Time Series Forecasting using Generative Adversarial Networks
• Continuous Conditional Generative Adversarial Networks (cGAN) with Generator Regularization
• Generation of Synthetic Electronic Health Records Using a Federated GAN
• Generation High resolution 3D model from natural language by Generative Adversarial Network
• Improving Federated Relational Data Modeling via Basis Alignment and Weight Penalty
• Generalized Energy Based Models
• Copy the Old or Paint Anew? An Adversarial Framework for (non-) Parametric Image Stylization
• Deep 3D-Zoom Net: Unsupervised Learning of Photo-Realistic 3D-Zoom
• No Surprises: Training Robust Lung Nodule Detection for Low-Dose CT Scans by Augmenting with Adversarial Attacks
• Data-Free Knowledge Distillation with Soft Targeted Transfer Set Synthesis
• Feature Affinity based Pseudo Labeling for Semi-supervised Person Re-identification
• Hyperprior Induced Unsupervised Disentanglement of Latent Representations
• PGMAN: An Unsupervised Generative Multi-adversarial Network for Pan-sharpening
• Unpaired Multi-Domain Image Generation via Regularized Conditional GANs
• EnvGAN: Adversarial Synthesis of Environmental Sounds for Data Augmentation
• Measuring Fairness in Generative Models
• Tree-Wasserstein Barycenter for Large-Scale Multilevel Clustering and Scalable Bayes
• Adversarial Regression. Generative Adversarial Networks for Non-Linear Regression: Theory and Assessment
• DISSECT: Disentangled Simultaneous Explanations via Concept Traversals
• FH-GAN: Face Hallucination and Recognition using Generative Adversarial Network
• Generative Adversarial Forests for Better Conditioned Adversarial Learning
• Extractive Summary as Discrete Latent Variables
• Lead Sheet Generation and Arrangement by Conditional Generative Adversarial Network
• Score-Guided Generative Adversarial Networks
• Ebola Optimization Search Algorithm (EOSA): A new metaheuristic algorithm based on the propagation model of Ebola virus disease
• Image-based Plant Disease Diagnosis with Unsupervised Anomaly Detection Based on Reconstructability of Colors
• SDE approximations of GANs training and its long-run behavior
• Gradient Layer: Enhancing the Convergence of Adversarial Training for Generative Models
• BrainNetGAN: Data augmentation of brain connectivity using generative adversarial network for dementia classification
• SRR-Net: A Super-Resolution-Involved Reconstruction Method for High Resolution MR Imaging
• Bayesian Reasoning with Trained Neural Networks
• Matrix Normal PCA for Interpretable Dimension Reduction and Graphical Noise Modeling
• Modeling EEG data distribution with a Wasserstein Generative Adversarial Network to predict RSVP Events
• Generative Adversarial Mapping Networks
• Boosting Image Outpainting with Semantic Layout Prediction
• Deep Convolutional GANs for Car Image Generation
• Variational Diffusion Autoencoders with Random Walk Sampling
• Extreme Image Coding via Multiscale Autoencoders With Generative Adversarial Optimization
• Fixed smooth convolutional layer for avoiding checkerboard artifacts in CNNs
• Reconstruction of Simulation-Based Physical Field by Reconstruction Neural Network Method
• Product-Aware Answer Generation in E-Commerce Question-Answering
• Hidden Convexity of Wasserstein GANs: Interpretable Generative Models with Closed-Form Solutions
• Learning Disentangled Representations of Satellite Image Time Series
• FaceShapeGene: A Disentangled Shape Representation for Flexible Face Image Editing
• Pipeline Generative Adversarial Networks for Facial Images Generation with Multiple Attributes
• Adversarial Constraint Learning for Structured Prediction
• Sparsely Grouped Multi-task Generative Adversarial Networks for Facial Attribute Manipulation
• CapsGAN: Using Dynamic Routing for Generative Adversarial Networks
• Correcting Nuisance Variation using Wasserstein Distance
• Manifold Interpolation for Large-Scale Multi-Objective Optimization via Generative Adversarial Networks
• Kernel-Based Training of Generative Networks
• Sliced Wasserstein Generative Models
• TET-GAN: Text Effects Transfer via Stylization and Destylization
• DotFAN: A Domain-transferred Face Augmentation Network for Pose and Illumination Invariant Face Recognition
• Learning Diverse Stochastic Human-Action Generators by Learning Smooth Latent Transitions
• Discriminative Active Learning for Domain Adaptation
• Few-shot Continual Learning: a Brain-inspired Approach
• On the Convergence of Gradient Descent in GANs: MMD GAN As a Gradient Flow
• Manifold-preserved GANs
• EVGen: Adversarial Networks for Learning Electric Vehicle Charging Loads and Hidden Representations
• SoftDICE for Imitation Learning: Rethinking Off-policy Distribution Matching
• Double Descent and Other Interpolation Phenomena in GANs
• The use of Generative Adversarial Networks to characterise new physics in multi-lepton final states at the LHC
• Imaginative Walks: Generative Random Walk Deviation Loss for Improved Unseen Learning Representation
• Self-Sampling for Neural Point Cloud Consolidation
• Large-Scale Optimal Transport via Adversarial Training with Cycle-Consistency
• A Novel Framework for Selection of GANs for an Application
• FIS-GAN: GAN with Flow-based Importance Sampling
• Stable Rank Normalization for Improved Generalization in Neural Networks and GANs
• Curriculum Learning for Deep Generative Models with Clustering
• Generative-Discriminative Complementary Learning
• Learning More with Less: Conditional PGGAN-based Data Augmentation for Brain Metastases Detection Using Highly-Rough Annotation on MR Images
• Parametric Synthesis of Text on Stylized Backgrounds using PGGANs
• On Lipschitz Bounds of General Convolutional Neural Networks
• Denoising of 3-D Magnetic Resonance Images Using a Residual Encoder-Decoder Wasserstein Generative Adversarial Network
• Emulating malware authors for proactive protection using GANs over a distributed image visualization of dynamic file behavior
• Adversarially Learned Mixture Model
• Risk Bounds for Unsupervised Cross-Domain Mapping with IPMs
• Batch Normalization in the final layer of generative networks
• Generative Visual Rationales
• MTGAN: Speaker Verification through Multitasking Triplet Generative Adversarial Networks
• Weakly Supervised Audio Source Separation via Spectrum Energy Preserved Wasserstein Learning
• Visual Analogies between Atari Games for Studying Transfer Learning in RL
• Text-to-image Synthesis via Symmetrical Distillation Networks
• Classification of sparsely labeled spatio-temporal data through semi-supervised adversarial learning
• Generating Diverse Programs with Instruction Conditioned Reinforced Adversarial Learning
• Synthetic Dynamic PMU Data Generation: A Generative Adversarial Network Approach
• Generating Realistic Sequences of Customer-level Transactions for Retail Datasets
• Defo-Net: Learning Body Deformation using Generative Adversarial Networks
• Wasserstein GAN Can Perform PCA
• Adaptive Wasserstein Hourglass for Weakly Supervised Hand Pose Estimation from Monocular RGB
• Language Modeling with Generative Adversarial Networks
• Adversarial Binary Coding for Efficient Person Re-identification
• Posterior asymptotics in Wasserstein metrics on the real line
• Semi-Supervised Learning with IPM-based GANs: an Empirical Study
• Generative Spatiotemporal Modeling Of Neutrophil Behavior
• Updating the generator in PPGN-h with gradients flowing through the encoder
• SparseGAN: Sparse Generative Adversarial Network for Text Generation
• On reproduction of On the regularization of Wasserstein GANs
• LDC-VAE: A Latent Distribution Consistency Approach to Variational AutoEncoders
• Resampling Base Distributions of Normalizing Flows
• Speech Enhancement-assisted Voice Conversion in Noisy Environments
• OCT-GAN: Neural ODE-based Conditional Tabular GANs
• Fast Computational Ghost Imaging using Unpaired Deep Learning and a Constrained Generative Adversarial Network
• Bitewing Radiography Semantic Segmentation Base on Conditional Generative Adversarial Nets
• An Improved StarGAN for Emotional Voice Conversion: Enhancing Voice Quality and Data Augmentation
• TableGAN-MCA: Evaluating Membership Collisions of GAN-Synthesized Tabular Data Releasing
• A Survey on GAN Acceleration Using Memory Compression Technique
• Does Adversarial Oversampling Help us?
• Protecting Anonymous Speech: A Generative Adversarial Network Methodology for Removing Stylistic Indicators in Text
• Data-driven Accelerogram Synthesis using Deep Generative Models
• Robust Attacks on Deep Learning Face Recognition in the Physical World
• GUIGAN: Learning to Generate GUI Designs Using Generative Adversarial Networks
• A Statistician Teaches Deep Learning
• Wasserstein Introspective Neural Networks
• An Iterative Closest Points Approach to Neural Generative Models
• End-to-end Generative Zero-shot Learning via Few-shot Learning
• OntoZSL: Ontology-enhanced Zero-shot Learning
• CIZSL++: Creativity Inspired Generative Zero-Shot Learning
• Deblurring Processor for Motion-Blurred Faces Based on Generative Adversarial Networks
• SETGAN: Scale and Energy Trade-off GANs for Image Applications on Mobile Platforms
• Unsupervised Learning of Solutions to Differential Equations with Generative Adversarial Networks
• Semi-supervised mp-MRI Data Synthesis with StitchLayer and Auxiliary Distance Maximization
• Out-of-Sample Testing for GANs
• f-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning
• LBS Autoencoder: Self-supervised Fitting of Articulated Meshes to Point Clouds
• Capturing the diversity of biological tuning curves using generative adversarial networks
• Implicit Manifold Learning on Generative Adversarial Networks
• Deep PDF: Probabilistic Surface Optimization and Density Estimation
• Wasserstein GAN and Waveform Loss-based Acoustic Model Training for Multi-speaker Text-to-Speech Synthesis Systems Using a WaveNet Vocoder
• LoGAN: Generating Logos with a Generative Adversarial Neural Network Conditioned on color
• Adversarial Learning of Label Dependency: A Novel Framework for Multi-class Classification
• Guiding the One-to-one Mapping in CycleGAN via Optimal Transport
• Conditional Transfer with Dense Residual Attention: Synthesizing traffic signs from street-view imagery
• Towards Language Agnostic Universal Representations
• CanvasGAN: A simple baseline for text to image generation by incrementally patching a canvas
• Local Stability and Performance of Simple Gradient Penalty mu-Wasserstein GAN
• Point Cloud Colorization Based on Densely Annotated 3D Shape Dataset
• Disassembling Object Representations without Labels
• Learning Interpretable and Discrete Representations with Adversarial Training for Unsupervised Text Classification
• Learning Energy-Based Approximate Inference Networks for Structured Applications in NLP
• LR-to-HR Face Hallucination with an Adversarial Progressive Attribute-Induced Network
• Shape-conditioned Image Generation by Learning Latent Appearance Representation from Unpaired Data
• Attending Category Disentangled Global Context for Image Classification
• An investigation of pre-upsampling generative modelling and Generative Adversarial Networks in audio super resolution
• Bridging the Gap between Label- and Reference-based Synthesis in Multi-attribute Image-to-Image Translation
• LSC-GAN: Latent Style Code Modeling for Continuous Image-to-image Translation
• High Diversity Attribute Guided Face Generation with GANs
• Learning convex regularizers satisfying the variational source condition for inverse problems
• What and Where to Translate: Local Mask-based Image-to-Image Translation
• Iterative Update and Unified Representation for Multi-Agent Reinforcement Learning
• EarthGAN: Can we visualize the Earth's mantle convection using a surrogate model?
• Deep Direct Likelihood Knockoffs
• Single Frame Deblurring with Laplacian Filters
• Adversarial training for predictive tasks: theoretical analysis and limitations in the deterministic case
• SimuGAN: Unsupervised forward modeling and optimal design of a LIDAR Camera
• Motion Generation Considering Situation with Conditional Generative Adversarial Networks for Throwing Robots
• Can adversarial training learn image captioning ?
• Anonymization of labeled TOF-MRA images for brain vessel segmentation using generative adversarial networks
• Image Translation for Medical Image Generation -- Ischemic Stroke Lesions
• Moreau-Yosida $f$-divergences
• Coupling Matrix Manifolds and Their Applications in Optimal Transport
• Summable Reparameterizations of Wasserstein Critics in the One-Dimensional Setting
• Physics Informed Convex Artificial Neural Networks (PICANNs) for Optimal Transport based Density Estimation
• A Frequency Domain Constraint for Synthetic and Real X-ray Image Super Resolution
• Ambient Hidden Space of Generative Adversarial Networks
• Instance Map based Image Synthesis with a Denoising Generative Adversarial Network
• Unsupervised Meta-learning of Figure-Ground Segmentation via Imitating Visual Effects
• Video-to-Video Translation for Visual Speech Synthesis
• Molecular Attributes Transfer from Non-Parallel Data
• Synthesizing Photorealistic Images with Deep Generative Learning
• Diversity Regularized Adversarial Learning
• Zero-Shot Learning Based on Knowledge Sharing
• NP-DRAW: A Non-Parametric Structured Latent Variable Model for Image Generation
• ScRAE: Deterministic Regularized Autoencoders with Flexible Priors for Clustering Single-cell Gene Expression Data
• Generating the support with extreme value losses
• Quality-aware Unpaired Image-to-Image Translation
• Lesion-Inspired Denoising Network: Connecting Medical Image Denoising and Lesion Detection
• NeuralDP Differentially private neural networks by design