A Note on the Convergence of Denoising Diffusion Probabilistic Models
This work provides theoretical guarantees for diffusion models, addressing convergence issues for arbitrary data distributions, which is incremental but important for the machine learning community.
The authors derived a quantitative upper bound on the Wasserstein distance between the data-generating distribution and the distribution learned by a diffusion model, without assumptions on the learned score function and avoiding exponential dependencies.
Diffusion models are one of the most important families of deep generative models. In this note, we derive a quantitative upper bound on the Wasserstein distance between the data-generating distribution and the distribution learned by a diffusion model. Unlike previous works in this field, our result does not make assumptions on the learned score function. Moreover, our bound holds for arbitrary data-generating distributions on bounded instance spaces, even those without a density w.r.t. the Lebesgue measure, and the upper bound does not suffer from exponential dependencies. Our main result builds upon the recent work of Mbacke et al. (2023) and our proofs are elementary.