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.
@article{arxiv.2408.04221,
title = {Connective Viewpoints of Signal-to-Noise Diffusion Models},
author = {Khanh Doan and Long Tung Vuong and Tuan Nguyen and Anh Tuan Bui and Quyen Tran and Thanh-Toan Do and Dinh Phung and Trung Le},
journal= {arXiv preprint arXiv:2408.04221},
year = {2024}
}