IID-GAN: an IID Sampling Perspective for Regularizing Mode Collapse

Abstract

Despite its success, generative adversarial networks (GANs) still suffer from mode collapse, i.e., the generator can only map latent variables to a partial set of modes in the target distribution. In this paper, we analyze and seek to regularize this issue with an independent and identically distributed (IID) sampling perspective and emphasize that holding the IID property referring to the target distribution for generation can naturally avoid mode collapse. This is based on the basic IID assumption for real data in machine learning. However, though the source samples {z} obey IID, the generations {G(z)} may not necessarily be IID sampling from the target distribution. Based on this observation, considering a necessary condition of IID generation that the inverse samples from target data should also be IID in the source distribution, we propose a new loss to encourage the closeness between inverse samples of real data and the Gaussian source in latent space to regularize the generation to be IID from the target distribution. Experiments on both synthetic and real-world data show the effectiveness of our model.

arXiv commentAccepted in IJCAI 2023

References in corpus (13)

• Auto-Encoding Variational Bayes
• Conditional Generative Adversarial Nets
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Spectral Normalization for Generative Adversarial Networks
• Variational Inference with Normalizing Flows
• Disentangling by Factorising
• VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
• Unrolled Generative Adversarial Networks
• Large Scale Adversarial Representation Learning
• Variational Approaches for Auto-Encoding Generative Adversarial Networks
• On GANs and GMMs
• Generative Multi-Adversarial Networks
• Prescribed Generative Adversarial Networks