Connective Viewpoints of Signal-to-Noise Diffusion Models

TL;DR

This paper provides a comprehensive analysis of Signal-to-Noise diffusion models, focusing on noise schedulers and their relation to SNR and information theory, and introduces a generalized backward equation to improve inference.

Contribution

It offers a unified perspective on noise schedulers in S2N diffusion models and proposes a new backward equation for better inference performance.

Findings

Enhanced inference with the generalized backward equation

Deeper understanding of noise schedulers via SNR and information theory

Improved performance across diffusion model applications

Abstract

Diffusion models (DM) have become fundamental components of generative models, excelling across various domains such as image creation, audio generation, and complex data interpolation. Signal-to-Noise diffusion models constitute a diverse family covering most state-of-the-art diffusion models. While there have been several attempts to study Signal-to-Noise (S2N) diffusion models from various perspectives, there remains a need for a comprehensive study connecting different viewpoints and exploring new perspectives. In this study, we offer a comprehensive perspective on noise schedulers, examining their role through the lens of the signal-to-noise ratio (SNR) and its connections to information theory. Building upon this framework, we have developed a generalized backward equation to enhance the performance of the inference process.