Improved Convergence Rate for Diffusion Probabilistic Models

Score-based diffusion models have demonstrated outstanding empirical performance in machine learning and artificial intelligence, particularly in generating high-quality new samples from complex probability distributions. Improving the theoretical understanding of diffusion models, with a particular focus on the convergence analysis, has attracted significant attention. In this work, we develop a convergence rate that is adaptive to the smoothness of different target distributions, referred to as instance-dependent bound. Specifically, we establish an iteration complexity of mind, d 2/3 L 1/3 , d 1/3 Lε -2/3 (up to logarithmic factors), where d denotes the data dimension, and ε quantifies the output accuracy in terms of total variation (TV) distance. In addition, L represents a relaxed Lipschitz constant, which, in the case of Gaussian mixture models, scales only logarithmically with the number of components, the dimension and iteration number, demonstrating broad applicability.