Stabilizing Generative Adversarial Networks: A Survey

2019/09/30 by Wiatrak, Maciej, Albrecht, Stefano V., Nystrom, Andrew · 10 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.1910.00927

Abstract

Generative Adversarial Networks (GANs) are a type of generative model which have received much attention due to their ability to model complex real-world data. Despite their recent successes, the process of training GANs remains challenging, suffering from instability problems such as non-convergence, vanishing or exploding gradients, and mode collapse. In recent years, a diverse set of approaches have been proposed which focus on stabilizing the GAN training procedure. The purpose of this survey is to provide a comprehensive overview of the GAN training stabilization methods which can be found in the literature. We discuss the advantages and disadvantages of each approach, offer a comparative summary, and conclude with a discussion of open problems.

Cited by

• Tackling Incomplete Data in Air Quality Prediction: A Bayesian Deep Learning Framework for Uncertainty Quantification
• Limited-Memory LRSGA: An Iterative Method for Computing Nash Equilibria in Competitive Optimization Problems
• A generative adversarial network optimization method for damage detection and digital twinning by deep AI fault learning: Z24 Bridge structural health monitoring benchmark validation
• An Encoder-Decoder Network for Beamforming over Sparse Large-Scale MIMO Channels
• RoentMod: A Synthetic Chest X-Ray Modification Model to Identify and Correct Image Interpretation Model Shortcuts
• CloudBreaker: Breaking the Cloud Covers of Sentinel-2 Images using Multi-Stage Trained Conditional Flow Matching on Sentinel-1
• Guided Unconditional and Conditional Generative Models for Super-Resolution and Inference of Quasi-Geostrophic Turbulence
• Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows
• Machine-Learning-Assisted Photonic Device Development: A Multiscale Approach from Theory to Characterization
• Adversarial Disentanglement by Backpropagation with Physics-Informed Variational Autoencoder

Related

• A Survey on Generative Adversarial Networks: Variants, Applications, and\n Training
• Unrolled Generative Adversarial Networks
• Smoothness and Stability in GANs
• Towards Principled Methods for Training Generative Adversarial Networks
• A New Perspective on Stabilizing GANs training: Direct Adversarial Training