Monte Carlo guided Diffusion for Bayesian linear inverse problems

Abstract

Ill-posed linear inverse problems arise frequently in various applications, from computational photography to medical imaging. A recent line of research exploits Bayesian inference with informative priors to handle the ill-posedness of such problems. Amongst such priors, score-based generative models (SGM) have recently been successfully applied to several different inverse problems. In this study, we exploit the particular structure of the prior defined by the SGM to define a sequence of intermediate linear inverse problems. As the noise level decreases, the posteriors of these inverse problems get closer to the target posterior of the original inverse problem. To sample from this sequence of posteriors, we propose the use of Sequential Monte Carlo (SMC) methods. The proposed algorithm, MCGDiff, is shown to be theoretically grounded and we provide numerical simulations showing that it outperforms competing baselines when dealing with ill-posed inverse problems in a Bayesian setting.

arXiv commentpreprint

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• Denoising Diffusion Probabilistic Models
• An Introduction to Variational Autoencoders
• The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo
• Deep Unsupervised Learning using Nonequilibrium Thermodynamics
• Score-Based Generative Modeling through Stochastic Differential Equations
• Normalizing Flows: An Introduction and Review of Current Methods
• Denoising Diffusion Restoration Models
• Diffusion Posterior Sampling for General Noisy Inverse Problems
• Solving Inverse Problems in Medical Imaging with Score-Based Generative Models
• Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problem
• Variational Autoencoder with Arbitrary Conditioning
• The Bayesian Approach To Inverse Problems
• Deep Unfolding with Normalizing Flow Priors for Inverse Problems
• Towards Coherent Image Inpainting Using Denoising Diffusion Implicit Models
• Practical and Asymptotically Exact Conditional Sampling in Diffusion Models
• Utilizing variational autoencoders in the Bayesian inverse problem of photoacoustic tomography