# SetGAN: Improving the stability and diversity of generative models through a permutation invariant architecture

## TL;DR

SetGAN introduces a permutation invariant architecture for GANs that enhances training stability and diversity, effectively addressing mode collapse and providing a new evaluation metric for generative models.

## Contribution

The paper proposes SetGAN, a novel permutation invariant architecture for GANs that improves stability, diversity, and introduces a new evaluation metric.

## Key findings

- SetGAN outperforms similar GAN variants in modeling accuracy.
- SetGAN is less sensitive to hyperparameter tuning.
- The new metric effectively evaluates GAN performance without prior application knowledge.

## Abstract

Generative adversarial networks (GANs) have proven effective in modeling distributions of high-dimensional data. However, their training instability is a well-known hindrance to convergence, which results in practical challenges in their applications to novel data. Furthermore, even when convergence is reached, GANs can be affected by mode collapse, a phenomenon for which the generator learns to model only a small part of the target distribution, disregarding the vast majority of the data manifold or distribution. This paper addresses these challenges by introducing SetGAN, an adversarial architecture that processes sets of generated and real samples, and discriminates between the origins of these sets (i.e., training versus generated data) in a flexible, permutation invariant manner. We also propose a new metric to quantitatively evaluate GANs that does not require previous knowledge of the application, apart from the data itself. Using the new metric, in conjunction with the state-of-the-art evaluation methods, we show that the proposed architecture, when compared with GAN variants stemming from similar strategies, produces more accurate models of the input data in a way that is also less sensitive to hyperparameter settings.