Residual-based error correction for neural operator accelerated infinite-dimensional Bayesian inverse problems

Abstract

We explore using neural operators, or neural network representations of nonlinear maps between function spaces, to accelerate infinite-dimensional Bayesian inverse problems (BIPs) with models governed by nonlinear parametric partial differential equations (PDEs). Neural operators have gained significant attention in recent years for their ability to approximate the parameter-to-solution maps defined by PDEs using as training data solutions of PDEs at a limited number of parameter samples. The computational cost of BIPs can be drastically reduced if the large number of PDE solves required for posterior characterization are replaced with evaluations of trained neural operators. However, reducing error in the resulting BIP solutions via reducing the approximation error of the neural operators in training can be challenging and unreliable. We provide an a priori error bound result that implies certain BIPs can be ill-conditioned to the approximation error of neural operators, thus leading to inaccessible accuracy requirements in training. To reliably deploy neural operators in BIPs, we consider a strategy for enhancing the performance of neural operators, which is to correct the prediction of a trained neural operator by solving a linear variational problem based on the PDE residual. We show that a trained neural operator with error correction can achieve a quadratic reduction of its approximation error, all while retaining substantial computational speedups of posterior sampling when models are governed by highly nonlinear PDEs. The strategy is applied to two numerical examples of BIPs based on a nonlinear reaction--diffusion problem and deformation of hyperelastic materials. We demonstrate that posterior representations of the two BIPs produced using trained neural operators are greatly and consistently enhanced by error correction.

References in corpus (20)

• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems
• Unsupervised discovery of interpretable hyperelastic constitutive laws
• A Comprehensive Survey of Neural Architecture Search: Challenges and Solutions
• Digital Image Correlation used to Analyze the Multiaxial Behavior of Rubber-like Materials
• Dimension-independent likelihood-informed MCMC
• Multipole Graph Neural Operator for Parametric Partial Differential Equations
• Scalable and efficient algorithms for the propagation of uncertainty from data through inference to prediction for large-scale problems, with application to flow of the Antarctic ice sheet
• Digital image correlation and finite element modelling as a method to determine mechanical properties of human soft tissue in vivo
• Data-Driven Model Reduction for the Bayesian Solution of Inverse Problems
• ResNet with one-neuron hidden layers is a Universal Approximator
• Accelerating MCMC with active subspaces
• Solving Large-Scale PDE-constrained Bayesian Inverse Problems with Riemann Manifold Hamiltonian Monte Carlo
• Bayesian Poroelastic Aquifer Characterization from InSAR Surface Deformation Data. Part I: Maximum A Posteriori Estimate
• Bayesian inference of heterogeneous epidemic models: Application to COVID-19 spread accounting for long-term care facilities
• Derivative-Informed Projected Neural Networks for High-Dimensional Parametric Maps Governed by PDEs
• A multiscale strategy for Bayesian inference using transport maps
• Multiplicative noise in Bayesian inverse problems: Well-posedness and consistency of MAP estimators
• Goal-Oriented A-Posteriori Estimation of Model Error as an Aid to Parameter Estimation
• Choosing observation operators to mitigate model error in Bayesian inverse problems

Cited by in corpus (9)

• Learning High-Dimensional Parametric Maps via Reduced Basis Adaptive Residual Networks
• DNN-MG: A Hybrid Neural Network/Finite Element Method with Applications to 3D Simulations of the Navier-Stokes Equations
• Neural Dynamical Operator: Continuous Spatial-Temporal Model with Gradient-Based and Derivative-Free Optimization Methods
• Neural Control of Discrete Weak Formulations: Galerkin, Least-Squares and Minimal-Residual Methods with Quasi-Optimal Weights
• Residual-Based Error Corrector Operator to Enhance Accuracy and Reliability of Neural Operator Surrogates of Nonlinear Variational Boundary-Value Problems
• MCMC-Net: Accelerating Markov Chain Monte Carlo with Neural Networks for Inverse Problems
• A Hybrid Two-level MCMC Framework to Accelerate Posterior Mean Estimation with Deep Learning Surrogates for Bayesian Inverse Problems
• Choosing observation operators to mitigate model error in Bayesian inverse problems
• Solver-in-the-Loop joint operator learning: fractional Laplace-Beltrami features for interface reconstruction