Abstract:The fundamental computational issues in Bayesian inverse problems (BIP) governed by partial differential equations (PDEs) stem from the requirement of repeated forward model evaluations. A popular strategy to reduce such costs is to replace expensive model simulations with computationally efficient approximations using operator learning, motivated by recent progress in deep learning. However, using the approximated model directly may introduce a modeling error, exacerbating the already ill-posedness of inverse problems. Thus, balancing between accuracy and efficiency is essential for the effective implementation of such approaches. To this end, we develop an adaptive operator learning framework that can reduce modeling error gradually by forcing the surrogate to be accurate in local areas. This is accomplished by adaptively fine-tuning the pre-trained approximate model with training points chosen by a greedy algorithm during the posterior evaluation process. To validate our approach, we use DeepOnet to construct the surrogate and unscented Kalman inversion (UKI) to approximate the BIP solution, respectively. Furthermore, we present a rigorous convergence guarantee in the linear case using the UKI framework. The approach is tested on a number of benchmarks, including the Darcy flow, the heat source inversion problem, and the reaction-diffusion problem. The numerical results show that our method can significantly reduce computational costs while maintaining inversion accuracy.

A Surrogate Hyperplane Bregman–Kaczmarz Method for Solving Linear Inverse Problems

A surrogate hyperplane Kaczmarz method for solving consistent linear equations

Convergence of Projected Subgradient Method with Sparse or Low-Rank Constraints

Adaptive Bregman-Kaczmarz: An Approach to Solve Linear Inverse Problems with Independent Noise Exactly

An accelerated Homotopy-Perturbation-Kaczmarz method for solving nonlinear inverse problems

Surrogate-Based Inverse Modeling of the Hydrological System: An Adaptive Approach Considering Surrogate Structural Error.

Quantile-based Random Sparse Kaczmarz for Corrupted, Noisy Linear Inverse Systems

Adaptive construction of surrogates for the Bayesian solution of inverse problems

The sparse Kaczmarz method with surrogate hyperplane for the regularized basis pursuit problem

A Fast Averaged Kaczmarz Iteration with Convex Penalty for Inverse Problems in Hilbert Spaces

Adaptive operator learning for infinite-dimensional Bayesian inverse problems

An Accelerated Kaczmarz Type Method for Nonlinear Inverse Problems in Banach Spaces with Uniformly Convex Penalty

A Projected Homotopy Perturbation Method for Nonlinear Inverse Problems in Banach Spaces

Preasymptotic convergence of randomized Kaczmarz method

Filter Based Methods for Statistical Linear Inverse Problems

Fast Subspace Optimization Method for Nonlinear Inverse Problems in Banach Spaces with Uniformly Convex Penalty Terms

On adaptive stochastic extended iterative methods for solving least squares

Solving Inverse Obstacle Scattering Problem with Latent Surface Representations

Efficient generalized Golub-Kahan based methods for dynamic inverse problems

Stochastic Data-Driven Bouligand Landweber Method for Solving Non-smooth Inverse Problems