Stochastic Deconvolutional Neural Network Ensemble Training on Generative Pseudo-Adversarial Networks

TL;DR

This paper proposes stochastic deconvolutional neural network ensemble training to enhance the stability and performance of Generative Pseudo-Adversarial Networks, addressing common issues like oscillation and mode collapse.

Contribution

It introduces a stochastic ensembling method specifically designed for training GANs, aiming to improve stability and mitigate common training problems.

Findings

Stochastic ensembling improves training stability.

Method reduces mode collapse occurrences.

Enhanced convergence behavior observed.

Abstract

The training of Generative Adversarial Networks is a difficult task mainly due to the nature of the networks. One such issue is when the generator and discriminator start oscillating, rather than converging to a fixed point. Another case can be when one agent becomes more adept than the other which results in the decrease of the other agent's ability to learn, reducing the learning capacity of the system as a whole. Additionally, there exists the problem of Mode Collapse which involves the generators output collapsing to a single sample or a small set of similar samples. To train GANs a careful selection of the architecture that is used along with a variety of other methods to improve training. Even when applying these methods there is low stability of training in relation to the parameters that are chosen. Stochastic ensembling is suggested as a method for improving the stability while training GANs.