arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.03332stat.MLcs.LG
收藏

一种基于混合模型的扩散模型引导框架

A Mixture-Based Framework for Guiding Diffusion Models

KTH University(KTH大学)

Yazid Janati, Badr Moufad, Mehdi Abou El Qassime, Alain Durmus, Eric Moulines, Jimmy Olsson

AI总结:

本文提出一种基于混合模型的扩散模型引导框架,通过混合近似中间后验分布并结合吉布斯采样,解决贝叶斯反问题中似然函数难以处理的问题,并在图像反问题和音频源分离任务上验证了有效性。

摘要:

去噪扩散模型在贝叶斯反问题领域推动了显著进展。近期方法利用预训练扩散模型作为先验来求解各类此类问题,仅使用推理时计算,从而无需在同一数据集上重新训练任务专用模型。为近似贝叶斯反问题的后验,扩散模型从一系列中间后验分布中采样,每个分布都具有难以处理的似然函数。本文针对这些中间分布提出了一种新颖的混合近似方法。由于这些混合分布中存在难以处理的项,直接基于梯度的采样不可行,因此我们提出了一种基于吉布斯采样的实用方法。我们通过在图像反问题上利用像素空间和潜在空间扩散先验进行大量实验,并在音频扩散模型的源分离任务上验证了所提方法。代码可在 https://www.github.com/badr-moufad/mgdm 获取。

英文摘要:

Denoising diffusion models have driven significant progress in the field of Bayesian inverse problems. Recent approaches use pre-trained diffusion models as priors to solve a wide range of such problems, only leveraging inference-time compute and thereby eliminating the need to retrain task-specific models on the same dataset. To approximate the posterior of a Bayesian inverse problem, a diffusion model samples from a sequence of intermediate posterior distributions, each with an intractable likelihood function. This work proposes a novel mixture approximation of these intermediate distributions. Since direct gradient-based sampling of these mixtures is infeasible due to intractable terms, we propose a practical method based on Gibbs sampling. We validate our approach through extensive experiments on image inverse problems, utilizing both pixel- and latent-space diffusion priors, as well as on source separation with an audio diffusion model. The code is available at https://www.github.com/badr-moufad/mgdm

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑