<ul><li>Diffusion models, based on discretizations of stochastic differential equations, are known for their generative performance.</li><li>A simplified theoretical framework has been proposed to analyze Euler-Maruyama discretization of variance-preserving SDEs in Denoising Diffusion Probabilistic Models (DDPMs).</li><li>The study leverages Gröenwall's inequality to establish a convergence rate of O(1/T^1/2) under Lipschitz assumptions, simplifying previous proofs.</li><li>Experiments validate the theory, confirming the error scaling, effectiveness of discrete noise over Gaussian noise, and the impact of incorrect noise scaling on performance.</li></ul>

A Simple Analysis of Discretization Error in Diffusion Models

Discover more