BI-DCGAN: A Theoretically Grounded Bayesian Framework for Efficient and Diverse GANs

TL;DR

BI-DCGAN introduces a Bayesian framework to GANs, incorporating model uncertainty to improve diversity and robustness, backed by theoretical proof and extensive experimental validation.

Contribution

This paper presents the first theoretical proof that Bayesian modeling enhances GAN sample diversity and introduces BI-DCGAN, a scalable Bayesian extension of DCGAN.

Findings

BI-DCGAN produces more diverse outputs than standard DCGANs.

Theoretical proof links Bayesian modeling to increased diversity.

Experimental results confirm improved robustness and efficiency.

Abstract

Generative Adversarial Networks (GANs) are proficient at generating synthetic data but continue to suffer from mode collapse, where the generator produces a narrow range of outputs that fool the discriminator but fail to capture the full data distribution. This limitation is particularly problematic, as generative models are increasingly deployed in real-world applications that demand both diversity and uncertainty awareness. In response, we introduce BI-DCGAN, a Bayesian extension of DCGAN that incorporates model uncertainty into the generative process while maintaining computational efficiency. BI-DCGAN integrates Bayes by Backprop to learn a distribution over network weights and employs mean-field variational inference to efficiently approximate the posterior distribution during GAN training. We establishes the first theoretical proof, based on covariance matrix analysis, that Bayesian modeling enhances sample diversity in GANs. We validate this theoretical result through extensive experiments on standard generative benchmarks, demonstrating that BI-DCGAN produces more diverse and robust outputs than conventional DCGANs, while maintaining training efficiency. These findings position BI-DCGAN as a scalable and timely solution for applications where both diversity and uncertainty are critical, and where modern alternatives like diffusion models remain too resource-intensive.