Markov chain generative adversarial neural networks for solving Bayesian inverse problems in physics applications
From MaRDI portal
Publication:6052371
Abstract: In the context of solving inverse problems for physics applications within a Bayesian framework, we present a new approach, Markov Chain Generative Adversarial Neural Networks (MCGANs), to alleviate the computational costs associated with solving the Bayesian inference problem. GANs pose a very suitable framework to aid in the solution of Bayesian inference problems, as they are designed to generate samples from complicated high-dimensional distributions. By training a GAN to sample from a low-dimensional latent space and then embedding it in a Markov Chain Monte Carlo method, we can highly efficiently sample from the posterior, by replacing both the high-dimensional prior and the expensive forward map. We prove that the proposed methodology converges to the true posterior in the Wasserstein-1 distance and that sampling from the latent space is equivalent to sampling in the high-dimensional space in a weak sense. The method is showcased on two test cases where we perform both state and parameter estimation simultaneously. The approach is shown to be up to two orders of magnitude more accurate than alternative approaches while also being up to two orders of magnitude computationally faster, in multiple test cases, including the important engineering setting of detecting leaks in pipelines.
Recommendations
- Solution of physics-based Bayesian inverse problems with deep generative priors
- Resolution-independent generative models based on operator learning for physics-constrained Bayesian inverse problems
- A dimension-reduced variational approach for solving physics-based inverse problems using generative adversarial network priors and normalizing flows
- Surrogate modeling for Bayesian inverse problems based on physics-informed neural networks
- Solving Bayesian inverse problems from the perspective of deep generative networks
Cites work
- A multigrid multilevel Monte Carlo method for transport in the Darcy-Stokes system
- Adaptive Gaussian Process Approximation for Bayesian Inference with Expensive Likelihood Functions
- An invitation to statistics in Wasserstein space
- Automated solution of differential equations by the finite element method. The FEniCS book
- Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
- Data assimilation: methods, algorithms, and applications
- Data-driven computational methods. Parameter and operator estimations
- Ensemble Kalman inversion: mean-field limit and convergence analysis
- Equation of state calculations by fast computing machines
- Handbook of Markov Chain Monte Carlo
- scientific article; zbMATH DE number 5180707 (Why is no real title available?)
- Inverse acoustic and electromagnetic scattering theory.
- Inverse problems: a Bayesian perspective
- Limitations of polynomial chaos expansions in the Bayesian solution of inverse problems
- Machine learning for fluid mechanics
- Markov chain Monte Carlo. Stochastic simulation for Bayesian inference.
- Measure theory. Vol. I and II
- Monte Carlo sampling methods using Markov chains and their applications
- Non-intrusive reduced order modeling of nonlinear problems using neural networks
- Numerical methods for stochastic computations. A spectral method approach.
- On the local Lipschitz stability of Bayesian inverse problems
- Posterior consistency for Gaussian process approximations of Bayesian posterior distributions
- Probability theory and stochastic processes
- Superconvergent discontinuous Galerkin methods for second-order elliptic problems
- The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo
- Transform-based particle filtering for elliptic Bayesian inverse problems
Cited in
(5)- Solution of physics-based inverse problems using conditional generative adversarial networks with full gradient penalty
- A dimension-reduced variational approach for solving physics-based inverse problems using generative adversarial network priors and normalizing flows
- Resolution-independent generative models based on operator learning for physics-constrained Bayesian inverse problems
- Adaptive event-triggered secure control for networked control systems subject to deception and replay cyber-attacks
- Deep orthogonal decomposition: a continuously adaptive neural network approach to model order reduction of parametrized partial differential equations
This page was built for publication: Markov chain generative adversarial neural networks for solving Bayesian inverse problems in physics applications
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6052371)
