Adaptive Monte Carlo augmented with normalizing flows

TL;DR

This paper introduces an adaptive MCMC method that combines local moves with normalizing flows to efficiently sample from complex, multi-modal high-dimensional distributions, especially when prior data is unavailable.

Contribution

It develops a novel adaptive MCMC algorithm integrating normalizing flows for nonlocal transitions, with theoretical convergence analysis and demonstrated efficiency in high-dimensional, multi-modal sampling.

Findings

Accelerates sampling across large free energy barriers.

Achieves faster equilibration between metastable modes.

Provides theoretical guarantees of convergence.

Abstract

Many problems in the physical sciences, machine learning, and statistical inference necessitate sampling from a high-dimensional, multi-modal probability distribution. Markov Chain Monte Carlo (MCMC) algorithms, the ubiquitous tool for this task, typically rely on random local updates to propagate configurations of a given system in a way that ensures that generated configurations will be distributed according to a target probability distribution asymptotically. In high-dimensional settings with multiple relevant metastable basins, local approaches require either immense computational effort or intricately designed importance sampling strategies to capture information about, for example, the relative populations of such basins. Here we analyze an adaptive MCMC which augments MCMC sampling with nonlocal transition kernels parameterized with generative models known as normalizing flows. We focus on a setting where there is no preexisting data, as is commonly the case for problems in which MCMC is used. Our method uses: (i) a MCMC strategy that blends local moves obtained from any standard transition kernel with those from a generative model to accelerate the sampling and (ii) the data generated this way to adapt the generative model and improve its efficacy in the MCMC algorithm. We provide a theoretical analysis of the convergence properties of this algorithm, and investigate numerically its efficiency, in particular in terms of its propensity to equilibrate fast between metastable modes whose rough location is known \textit{a~priori} but respective probability weight is not. We show that our algorithm can sample effectively across large free energy barriers, providing dramatic accelerations relative to traditional MCMC algorithms.