HGAN: Hybrid Generative Adversarial Network

TL;DR

This paper introduces HGAN, a hybrid GAN model that combines adversarial training with explicit density estimation to prevent mode collapse and improve sample diversity, validated on multiple datasets.

Contribution

HGAN is the first to integrate autoregressive density estimation with GANs through adversarial knowledge transfer, enhancing mode coverage and sample quality.

Findings

HGAN outperforms baseline models on MNIST, CIFAR-10, STL-10.

HGAN effectively mitigates mode collapse in GAN training.

HGAN achieves superior quantitative and qualitative results.

Abstract

In this paper, we present a simple approach to train Generative Adversarial Networks (GANs) in order to avoid a \textit {mode collapse} issue. Implicit models such as GANs tend to generate better samples compared to explicit models that are trained on tractable data likelihood. However, GANs overlook the explicit data density characteristics which leads to undesirable quantitative evaluations and mode collapse. To bridge this gap, we propose a hybrid generative adversarial network (HGAN) for which we can enforce data density estimation via an autoregressive model and support both adversarial and likelihood framework in a joint training manner which diversify the estimated density in order to cover different modes. We propose to use an adversarial network to \textit {transfer knowledge} from an autoregressive model (teacher) to the generator (student) of a GAN model. A novel deep architecture within the GAN formulation is developed to adversarially distill the autoregressive model information in addition to simple GAN training approach. We conduct extensive experiments on real-world datasets (i.e., MNIST, CIFAR-10, STL-10) to demonstrate the effectiveness of the proposed HGAN under qualitative and quantitative evaluations. The experimental results show the superiority and competitiveness of our method compared to the baselines.