Tackling the Generative Learning Trilemma with Denoising Diffusion GANs

TL;DR

This paper introduces denoising diffusion GANs, a novel approach that significantly accelerates diffusion models by modeling denoising steps with multimodal GANs, achieving high sample quality and diversity at a fraction of the original sampling time.

Contribution

The paper proposes denoising diffusion GANs that reduce diffusion sampling time by modeling denoising steps with multimodal GANs, enabling practical real-world application.

Findings

Achieves 2000x faster sampling on CIFAR-10.

Maintains competitive sample quality and diversity.

Outperforms traditional GANs in mode coverage.

Abstract

A wide variety of deep generative models has been developed in the past decade. Yet, these models often struggle with simultaneously addressing three key requirements including: high sample quality, mode coverage, and fast sampling. We call the challenge imposed by these requirements the generative learning trilemma, as the existing models often trade some of them for others. Particularly, denoising diffusion models have shown impressive sample quality and diversity, but their expensive sampling does not yet allow them to be applied in many real-world applications. In this paper, we argue that slow sampling in these models is fundamentally attributed to the Gaussian assumption in the denoising step which is justified only for small step sizes. To enable denoising with large steps, and hence, to reduce the total number of denoising steps, we propose to model the denoising distribution using a complex multimodal distribution. We introduce denoising diffusion generative adversarial networks (denoising diffusion GANs) that model each denoising step using a multimodal conditional GAN. Through extensive evaluations, we show that denoising diffusion GANs obtain sample quality and diversity competitive with original diffusion models while being 2000$\times$ faster on the CIFAR-10 dataset. Compared to traditional GANs, our model exhibits better mode coverage and sample diversity. To the best of our knowledge, denoising diffusion GAN is the first model that reduces sampling cost in diffusion models to an extent that allows them to be applied to real-world applications inexpensively. Project page and code can be found at https://nvlabs.github.io/denoising-diffusion-gan