paper-with-me

홈 › Papers

Improved Convergence Rate for Diffusion Probabilistic Models

2024-10-17 · Gen Li, Yuchen Jiao

Score-based diffusion models have achieved remarkable empirical performance in the field of machine learning and artificial intelligence for their ability to generate high-quality new data instances from complex distributions. Improving our understanding of diffusion models, including mainly convergence analysis for such models, has attracted a lot of interests. Despite a lot of theoretical attempts, there still exists significant gap between theory and practice. Towards to close this gap, we establish an iteration complexity at the order of $d^{1/3}\varepsilon^{-2/3}$, which is better than $d^{5/12}\varepsilon^{-1}$, the best known complexity achieved before our work. This convergence analysis is based on a randomized midpoint method, which is first proposed for log-concave sampling (Shen and Lee, 2019), and then extended to diffusion models by Gupta et al. (2024). Our theory accommodates $\varepsilon$-accurate score estimates, and does not require log-concavity on the target distribution. Moreover, the algorithm can also be parallelized to run in only $O(\log^2(d/\varepsilon))$ parallel rounds in a similar way to prior works.

📄 PDF Abstract BibTeX arXiv:2410.13738

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

An overview of diffusion models for generative artificial intelligence

2024-12-02 · Davide Gallon, Arnulf Jentzen, Philippe von Wurstemberger

This article provides a mathematically rigorous introduction to denoising diffusion probabilistic models (DDPMs), sometimes also referred to as diffusion probabilistic models or diffusion models, for generative artificia…

Denoising

Improved Denoising Diffusion Probabilistic Models

2021-02-18 · Alex Nichol, Prafulla Dhariwal

Denoising diffusion probabilistic models (DDPM) are a class of generative models which have recently been shown to produce excellent samples. We show that with a few simple modifications, DDPMs can also achieve competiti…

DenoisingImage Generation

O(d/T) Convergence Theory for Diffusion Probabilistic Models under Minimal Assumptions

2024-09-27 · Gen Li, Yuling Yan

Score-based diffusion models, which generate new data by learning to reverse a diffusion process that perturbs data from the target distribution into noise, have achieved remarkable success across various generative task…

Denoising

Convergence Analysis for Entropy-Regularized Control Problems: A Probabilistic Approach

2024-06-16 · Jin Ma, Gaozhan Wang, Jianfeng Zhang

In this paper we investigate the convergence of the Policy Iteration Algorithm (PIA) for a class of general continuous-time entropy-regularized stochastic control problems. In particular, instead of employing sophisticat…

A Simple Analysis of Discretization Error in Diffusion Models

2025-06-10 · Juhyeok Choi, Chenglin Fan

Diffusion models, formulated as discretizations of stochastic differential equations (SDEs), achieve state-of-the-art generative performance. However, existing analyses of their discretization error often rely on complex…

Denoising