Provable Probabilistic Imaging using Score-Based Generative Priors
This work addresses the need for reliable imaging in inverse problems, offering a principled framework for practitioners, though it appears incremental as it builds on existing plug-and-play and denoising methods.
The paper tackles the problem of high-quality image reconstruction with uncertainty quantification in ill-posed inverse problems by proposing plug-and-play Monte Carlo (PMC) algorithms, which incorporate score-based generative priors and achieve improved reconstruction quality and high-fidelity uncertainty quantification in experiments.
Estimating high-quality images while also quantifying their uncertainty are two desired features in an image reconstruction algorithm for solving ill-posed inverse problems. In this paper, we propose plug-and-play Monte Carlo (PMC) as a principled framework for characterizing the space of possible solutions to a general inverse problem. PMC is able to incorporate expressive score-based generative priors for high-quality image reconstruction while also performing uncertainty quantification via posterior sampling. In particular, we develop two PMC algorithms that can be viewed as the sampling analogues of the traditional plug-and-play priors (PnP) and regularization by denoising (RED) algorithms. To improve the sampling efficiency, we introduce weighted annealing into these PMC algorithms, further developing two additional annealed PMC algorithms (APMC). We establish a theoretical analysis for characterizing the convergence behavior of PMC algorithms. Our analysis provides non-asymptotic stationarity guarantees in terms of the Fisher information, fully compatible with the joint presence of weighted annealing, potentially non-log-concave likelihoods, and imperfect score networks. We demonstrate the performance of the PMC algorithms on multiple representative inverse problems with both linear and nonlinear forward models. Experimental results show that PMC significantly improves reconstruction quality and enables high-fidelity uncertainty quantification.