Abstract:Original data usually lies on a set of disconnected manifolds rather than a smooth connected manifold. This causes the problem of mode collapse in the training of vanilla Generative Adversarial Network (GAN). There are many existing GAN variants that attempt to address this problem, but they result in limitations. The existing variants either produce simulated instances with low quality or generate identical simulated instances. In this study, we propose a new approach to training GAN utilizing multiple generators, a classifier and a discriminator to address mode collapse. The classifier outputs the statistical probabilities of generated data belonging to a specific category. These probabilities implicitly reflect which manifolds are captured by generators, and the correlation between generators is quantified by mutual information. Our idea views the mutual information values as a constraint to guide generators in learning different manifolds. Specifically, we traverse the generators, calculating the mutual information between each generator and the others. The calculated values are integrated into the generator loss to form a new generator loss and to update the corresponding generator's parameters, using back-propagation. We minimize the mutual information to reduce the correlation between generators while also minimizing the generator loss. This ensures generators capture different manifolds while updating their parameters. A new minimax formula is established to train our approach in a similar spirit to vanilla GAN. We term our approach Mutual Information Multi-generator GAN (MIM-GAN). We conduct extensive experiments utilizing the MNIST, CIFAR10 and CelebA datasets to demonstrate the significant performance improvement of MIM-GAN in both achieving the highest Inception Scores and producing diverse generated data at different resolutions.

Generative Adversarial Networks with Joint Distribution Moment Matching

Dual Distribution Matching GAN

Conditional Generative Moment-Matching Networks

MMD GAN: Towards Deeper Understanding of Moment Matching Network

JDGAN: Enhancing generator on extremely limited data via joint distribution

Generative Adversarial Mapping Networks

Improving Generative Moment Matching Networks with Distribution Partition

Partial Modal Conditioned GANs for Multi-modal Multi-label Learning with Arbitrary Modal-Missing

MMCGAN: Generative Adversarial Network with Explicit Manifold Prior

JGAN: A Joint Formulation of GAN for Synthesizing Images and Labels

Matching Attention Network For Domain Adaptation Optimized By Joint Gans And Kl-Mmd

Dual Discriminator Weighted Mixture Generative Adversarial Network for image generation

Multi-generator GAN learning disconnected manifolds with mutual information

MatchGAN: A Self-Supervised Semi-Supervised Conditional Generative Adversarial Network

Label-Removed Generative Adversarial Networks Incorporating with K-Means

Multi-distribution Mixture Generative Adversarial Networks for Fitting Diverse Data Sets

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators

Enhancing Stability in Training Conditional Generative Adversarial Networks via Selective Data Matching

DMGAN: Discriminative Metric-based Generative Adversarial Networks.

A Generative Adversarial Distribution Matching Framework for Visual Domain Adaptation

chi(2) Generative Adversarial Network