Semi-supervised invertible neural operators for Bayesian inverse problems
From MaRDI portal
Abstract: Neural Operators offer a powerful, data-driven tool for solving parametric PDEs as they can represent maps between infinite-dimensional function spaces. In this work, we employ physics-informed Neural Operators in the context of high-dimensional, Bayesian inverse problems. Traditional solution strategies necessitate an enormous, and frequently infeasible, number of forward model solves, as well as the computation of parametric derivatives. In order to enable efficient solutions, we extend Deep Operator Networks (DeepONets) by employing a RealNVP architecture which yields an invertible and differentiable map between the parametric input and the branch-net output. This allows us to construct accurate approximations of the full posterior, irrespective of the number of observations and the magnitude of the observation noise, without any need for additional forward solves nor for cumbersome, iterative sampling procedures. We demonstrate the efficacy and accuracy of the proposed methodology in the context of inverse problems for three benchmarks: an anti-derivative equation, reaction-diffusion dynamics and flow through porous media.
Recommendations
- Neural operator induced Gaussian process framework for probabilistic solution of parametric partial differential equations
- Adaptive operator learning for infinite-dimensional Bayesian inverse problems
- Physics-informed discretization-independent deep compositional operator network
- Operator learning with Gaussian processes
- Resolution-independent generative models based on operator learning for physics-constrained Bayesian inverse problems
Cites work
- A multi-resolution, non-parametric, Bayesian framework for identification of spatially-varying model parameters
- Automated solution of differential equations by the finite element method. The FEniCS book
- Bayesian model and dimension reduction for uncertainty propagation: applications in random media
- Data-driven model reduction and transfer operator approximation
- Deep learning models for global coordinate transformations that linearise PDEs
- Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
- Geometric MCMC for infinite-dimensional inverse problems
- Graphical models, exponential families, and variational inference
- Incorporating physical constraints in a deep probabilistic machine learning framework for coarse-graining dynamical systems
- Model reduction and neural networks for parametric PDEs
- Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders
- Multimodal, high-dimensional, model-based, Bayesian inverse problems with applications in biomechanics
- Pattern recognition and machine learning.
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Solving ill-posed inverse problems using iterative deep neural networks
- Special issue: Big data and predictive computational modeling
Cited in
(30)- Variational inference for nonlinear inverse problems via neural net kernels: comparison to Bayesian neural networks, application to topology optimization
- B-DeepONet: an enhanced Bayesian deeponet for solving noisy parametric PDEs using accelerated replica exchange SGLD
- On obtaining sparse semantic solutions for inverse problems, control, and neural network training
- Solving Inverse Problems by Joint Posterior Maximization with Autoencoding Prior
- Fully probabilistic deep models for forward and inverse problems in parametric PDEs
- Residual-based error correction for neural operator accelerated Infinite-dimensional Bayesian inverse problems
- AI‐enhanced iterative solvers for accelerating the solution of large‐scale parametrized systems
- Residual-based error corrector operator to enhance accuracy and reliability of neural operator surrogates of nonlinear variational boundary-value problems
- Resolution-independent generative models based on operator learning for physics-constrained Bayesian inverse problems
- DNN-MG: a hybrid neural network/finite element method with applications to 3D simulations of the Navier-Stokes equations
- Derivative-informed neural operator: an efficient framework for high-dimensional parametric derivative learning
- Mitigating spectral bias for the multiscale operator learning with hierarchical attention
- A MCMC method based on surrogate model and Gaussian process parameterization for infinite Bayesian PDE inversion
- Efficient Bayesian physics informed neural networks for inverse problems via ensemble Kalman inversion
- D2NO: efficient handling of heterogeneous input function spaces with distributed deep neural operators
- Out-of-distributional risk bounds for neural operators with applications to the Helmholtz equation
- Pseudo-differential integral autoencoder network for inverse PDE operators
- Neural operator induced Gaussian process framework for probabilistic solution of parametric partial differential equations
- Physics-informed discretization-independent deep compositional operator network
- Physics-informed multi-grid neural operator: theory and an application to porous flow simulation
- Weak neural variational inference for solving Bayesian inverse problems \textit{without} forward models: applications in elastography
- Physics-informed geometry-aware neural operator
- Operator learning with Gaussian processes
- Adaptive operator learning for infinite-dimensional Bayesian inverse problems
- Solving forward and inverse partial differential equation problems on unknown manifolds via physics-informed neural operators
- DGennO: a novel physics-aware neural operator for solving forward and inverse PDE problems based on deep, generative probabilistic modeling
- Physics-aware deep learning framework for the limited aperture inverse obstacle scattering problem
- How regularization terms make invertible neural networks Bayesian point estimators
- A data-driven method for modeling creep-fatigue stress-strain behavior using neural ODEs
- GenPANIS: a latent-variable generative framework for forward and inverse PDE problems in multiphase media
This page was built for publication: Semi-supervised invertible neural operators for Bayesian inverse problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6164274)
