Kinetic Langevin Meets Split Gibbs: Accelerated Posterior Sampling for Imaging Inverse Problems with Diffusion Priors

Key points

• Langevin-within-SGS takes cheap overdamped Langevin steps but needs many iterations.1
• We propose RED-KLwSGS, which keeps the exact Gaussian update for the data variable and updates the auxiliary variable with underdamped (kinetic) Langevin diffusions driven by a one-shot denoising score, at the same per-iteration cost as Langevin-within-SGS.1
• We also introduce Joint-RED-KLwSGS, which applies kinetic Langevin diffusions to both variables.1
• Experiments with Denoising diffusion probabilistic models as diffusion priors on FFHQ and ImageNet datasets show faster convergence and high-quality image reconstruction.1

Sources (1)

• [1]Kinetic Langevin Meets Split Gibbs: Accelerated Posterior Sampling for Imaging Inverse Problems with Diffusion Priors
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 02:53 PM

  Split Gibbs sampling (SGS) is a popular framework for posterior sampling in Bayesian imaging inverse problems.

  Langevin-within-SGS takes cheap overdamped Langevin steps but needs many iterations.

Extractive summary: sentences quoted from the sources.