IVE-GAN: Invariant Encoding Generative Adversarial Networks

TL;DR

IVE-GAN introduces an invariant encoding mechanism to GANs, enabling better mode coverage and richer data representations by mapping individual samples to the latent space using transformation-invariant features.

Contribution

The paper proposes IVE-GAN, a novel framework that adds an inverse mapping from data to latent space based on invariant features, improving mode coverage and representation quality.

Findings

Enhanced mode coverage in generated data

Improved representation learning on benchmark datasets

Effective in common image generation tasks

Abstract

Generative adversarial networks (GANs) are a powerful framework for generative tasks. However, they are difficult to train and tend to miss modes of the true data generation process. Although GANs can learn a rich representation of the covered modes of the data in their latent space, the framework misses an inverse mapping from data to this latent space. We propose Invariant Encoding Generative Adversarial Networks (IVE-GANs), a novel GAN framework that introduces such a mapping for individual samples from the data by utilizing features in the data which are invariant to certain transformations. Since the model maps individual samples to the latent space, it naturally encourages the generator to cover all modes. We demonstrate the effectiveness of our approach in terms of generative performance and learning rich representations on several datasets including common benchmark image generation tasks.