Complexity Controlled Generative Adversarial Networks

TL;DR

This paper introduces a novel approach to GAN training using Low-Complexity Neural Networks (LCNNs) to enhance stability and prevent mode collapse, demonstrated on large image datasets.

Contribution

The paper proposes integrating LCNN loss functions into GAN architectures to control model complexity and improve training stability and diversity.

Findings

Stable training achieved across multiple GAN variants.

Mode collapse is effectively avoided.

Increased inception scores indicate better generative quality.

Abstract

One of the issues faced in training Generative Adversarial Nets (GANs) and their variants is the problem of mode collapse, wherein the training stability in terms of the generative loss increases as more training data is used. In this paper, we propose an alternative architecture via the Low-Complexity Neural Network (LCNN), which attempts to learn models with low complexity. The motivation is that controlling model complexity leads to models that do not overfit the training data. We incorporate the LCNN loss function for GANs, Deep Convolutional GANs (DCGANs) and Spectral Normalized GANs (SNGANs), in order to develop hybrid architectures called the LCNN-GAN, LCNN-DCGAN and LCNN-SNGAN respectively. On various large benchmark image datasets, we show that the use of our proposed models results in stable training while avoiding the problem of mode collapse, resulting in better training stability. We also show how the learning behavior can be controlled by a hyperparameter in the LCNN functional, which also provides an improved inception score.