Unveil Conditional Diffusion Models with Classifier-free Guidance: A Sharp Statistical Theory

arXiv:2403.11968v1 Announce Type: new
Abstract: Conditional diffusion models serve as the foundation of modern image synthesis and find extensive application in fields like computational biology and reinforcement learning. In these applications, conditional diffusion models incorporate various conditional information, such as prompt input, to guide the sample generation towards desired properties. Despite the empirical success, theory of conditional diffusion models is largely missing. This paper bridges this gap by presenting a sharp statistical theory of distribution estimation using conditional diffusion models. …

abstract application applications arxiv biology classifier computational computational biology cs.lg diffusion diffusion models fields foundation free guidance guide image information math.st modern prompt reinforcement reinforcement learning sample serve statistical stat.ml stat.th synthesis theory type