Overcoming Mode Collapse with Adaptive Multi Adversarial Training
Not on the map. Why?
Abstract (arXiv)
Generative Adversarial Networks (GANs) are a class of generative models used for various applications, but they have been known to suffer from the mode collapse problem, in which some modes of the target distribution are ignored by the generator. Investigative study using a new data generation procedure indicates that the mode collapse of the generator is driven by the discriminator's inability to maintain classification accuracy on previously seen samples, a phenomenon called Catastrophic Forgetting in continual learning. Motivated by this observation, we introduce a novel training procedure that adaptively spawns additional discriminators to remember previous modes of generation. On several datasets, we show that our training scheme can be plugged-in to existing GAN frameworks to mitigate mode collapse and improve standard metrics for GAN evaluation.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 32 of its 47 references that are in the map. First posted on arXiv 29 Dec 2021.
Builds on
32 of 47 references.
References · 32
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
- 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
- 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
- 2016Improved Techniques for Training GANscited by 2,970
- 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
- 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,333
- 2017Improved Training of Wasserstein GANscited by 2,235
- 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
- 2018Spectral Normalization for Generative Adversarial Networkscited by 1,636
- 2016Progressive Neural Networkscited by 1,361
20 more
- 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,082
- 2018Which Training Methods for GANs do actually Converge?cited by 679
- 2017Wasserstein GANcited by 639
- 2017PathNet: Evolution Channels Gradient Descent in Super Neural Networkscited by 364
- 2020PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Modelscited by 358
- 2018Generative Adversarial Network in Medical Imaging: A Reviewcited by 257
- 2016Unrolled Generative Adversarial Networkscited by 208
- 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 200
- 2019Seeing What a GAN Cannot Generatecited by 148
- 2017Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Graspingcited by 144
- 2017PacGAN: The power of two samples in generative adversarial networkscited by 104
- 2020RL-CycleGAN: Reinforcement Learning Aware Simulation-To-Realcited by 91
- 2016Generative Multi-Adversarial Networkscited by 60
- 2017Dual Discriminator Generative Adversarial Netscited by 55
- 2017AdaGAN: Boosting Generative Modelscited by 34
- 2019Self-supervised GAN: Analysis and Improvement with Multi-class Minimax Gamecited by 25
- 2019Side-Tuning: A Baseline for Network Adaptation via Additive Side Networkscited by 16
- 2018Generative Adversarial Network Training is a Continual Learning Problemcited by 13
- 2020microbatchGAN: Stimulating Diversity with Multi-Adversarial Discriminationcited by 7
- 2018Self-Supervised GAN to Counter Forgettingcited by 2
Built on by
Citing papers per year: 2022: 1; 2023: 1; 2024: 1.
3 papers.