MCMC-Net: Accelerating Markov Chain Monte Carlo with Neural Networks for Inverse Problems

Abstract

In many computational problems, using the Markov Chain Monte Carlo (MCMC) can be prohibitively time-consuming. We propose MCMC-Net, a simple yet efficient way to accelerate MCMC via neural networks. The key idea of our approach is to substitute the true likelihood function of the MCMC method with a neural operator based surrogate. We extensively evaluate the accuracy and speedup of our method on three different PDE-based inverse problems where likelihood computations are computationally expensive, namely electrical impedance tomography, diffuse optical tomography, and quantitative photoacoustic tomography. MCMC-Net performs similar to the classical likelihood counterpart but with a significant speedup. We conjecture that the method can be applied to any problem with a sufficiently expensive likelihood function. We also analyze MCMC-Net in a theoretical setting for the different use cases. We prove a universal approximation theorem-type result to show that the proposed network can approximate the mapping resulting from forward model evaluations to a desired accuracy. Furthermore, we establish convergence of the surrogate posterior to the true posterior under Hellinger distance.

41 pages, 14 figures, 11 tables

References in corpus (8)

• FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators
• Learning Operators with Coupled Attention
• Convolutional Neural Operators for robust and accurate learning of PDEs
• Residual-based error correction for neural operator accelerated infinite-dimensional Bayesian inverse problems
• NOMAD: Nonlinear Manifold Decoders for Operator Learning
• Neural Inverse Operators for Solving PDE Inverse Problems
• Variable-Input Deep Operator Networks
• Globally injective and bijective neural operators