A Mixture-Based Framework for Guiding Diffusion Models
This work addresses a specific bottleneck in using diffusion models for inverse problems, offering an incremental improvement in sampling efficiency.
The paper tackles the challenge of approximating intractable intermediate posterior distributions in diffusion models for Bayesian inverse problems by proposing a novel mixture approximation and a practical Gibbs sampling method, achieving validation through extensive experiments on image and audio tasks.
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