DiffFlow: A Unified SDE Framework for Score-Based Diffusion Models and Generative Adversarial Networks

by   Jingwei Zhang, et al.

Generative models can be categorized into two types: explicit generative models that define explicit density forms and allow exact likelihood inference, such as score-based diffusion models (SDMs) and normalizing flows; implicit generative models that directly learn a transformation from the prior to the data distribution, such as generative adversarial nets (GANs). While these two types of models have shown great success, they suffer from respective limitations that hinder them from achieving fast sampling and high sample quality simultaneously. In this paper, we propose a unified theoretic framework for SDMs and GANs. We shown that: i) the learning dynamics of both SDMs and GANs can be described as a novel SDE named Discriminator Denoising Diffusion Flow (DiffFlow) where the drift can be determined by some weighted combinations of scores of the real data and the generated data; ii) By adjusting the relative weights between different score terms, we can obtain a smooth transition between SDMs and GANs while the marginal distribution of the SDE remains invariant to the change of the weights; iii) we prove the asymptotic optimality and maximal likelihood training scheme of the DiffFlow dynamics; iv) under our unified theoretic framework, we introduce several instantiations of the DiffFLow that provide new algorithms beyond GANs and SDMs with exact likelihood inference and have potential to achieve flexible trade-off between high sample quality and fast sampling speed.


page 1

page 2

page 3

page 4


Flow-GAN: Bridging implicit and prescribed learning in generative models

Evaluating the performance of generative models for unsupervised learnin...

Semi-Implicit Denoising Diffusion Models (SIDDMs)

Despite the proliferation of generative models, achieving fast sampling ...

GREAT Score: Global Robustness Evaluation of Adversarial Perturbation using Generative Models

Current studies on adversarial robustness mainly focus on aggregating lo...

Tackling the Generative Learning Trilemma with Denoising Diffusion GANs

A wide variety of deep generative models has been developed in the past ...

Stein Bridging: Enabling Mutual Reinforcement between Explicit and Implicit Generative Models

Deep generative models are generally categorized into explicit models an...

The Score-Difference Flow for Implicit Generative Modeling

Implicit generative modeling (IGM) aims to produce samples of synthetic ...

Can denoising diffusion probabilistic models generate realistic astrophysical fields?

Score-based generative models have emerged as alternatives to generative...

Please sign up or login with your details

Forgot password? Click here to reset