Derivative-informed projected neural networks for high-dimensional parametric maps governed by PDEs

DOI10.1016/j.cma.2021.114199OpenAlexW3205912750WikidataQ114196872 ScholiaQ114196872MaRDI QIDQ2060092

Thomas O'Leary-Roseberry, Peng Chen, Umberto Villa, Omar Ghattas

Publication date: 13 December 2021

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2011.15110

zbMATH Keywords

neural networksdeep learningsurrogate modelingactive subspaceparametrized PDEsderivative-informed dimension reduction

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) General nonlinear regression (62J02) Numerical methods for partial differential equations, initial value and time-dependent initial-boundary value problems (65M99) Neural nets and related approaches to inference from stochastic processes (62M45)

Related Items (13)

Projected Wasserstein Gradient Descent for High-Dimensional Bayesian Inference ⋮ An Offline-Online Decomposition Method for Efficient Linear Bayesian Goal-Oriented Optimal Experimental Design: Application to Optimal Sensor Placement ⋮ Learning physics-based models from data: perspectives from inverse problems and model reduction ⋮ Data-driven uncertainty quantification in computational human head models ⋮ slimTrain---A Stochastic Approximation Method for Training Separable Deep Neural Networks ⋮ Learning high-dimensional parametric maps via reduced basis adaptive residual networks ⋮ Sparse polynomial approximations for affine parametric saddle point problems ⋮ A Fast and Scalable Computational Framework for Large-Scale High-Dimensional Bayesian Optimal Experimental Design ⋮ Convergence Rates for Learning Linear Operators from Noisy Data ⋮ Multi‐fidelity data fusion through parameter space reduction with applications to automotive engineering ⋮ Large-scale Bayesian optimal experimental design with derivative-informed projected neural network ⋮ Learning the random variables in Monte Carlo simulations with stochastic gradient descent: Machine learning for parametric PDEs and financial derivative pricing ⋮ Data assimilation -- mathematical foundation and applications. Abstracts from the workshop held February 20--26, 2022

Uses Software

• TensorFlow
• hIPPYlib
• hIPPYflow

Cites Work

• Unnamed Item
• Unnamed Item
• An Explicit Link between Gaussian Fields and Gaussian Markov Random Fields: The Stochastic Partial Differential Equation Approach
• Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions
• Dimensionality reduction of parameter-dependent problems through proper orthogonal decomposition
• 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
• Mitigating the influence of the boundary on PDE-based covariance operators
• Optimal design of acoustic metamaterial cloaks under uncertainty
• Generalized bounds for active subspaces
• Taylor approximation and variance reduction for PDE-constrained optimal control under uncertainty
• Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
• A scalable algorithm for MAP estimators in Bayesian inverse problems with Besov priors
• Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems
• A new algorithm for high-dimensional uncertainty quantification based on dimension-adaptive sparse grid approximation and reduced basis methods
• Analysis of the Hessian for inverse scattering problems. III: Inverse medium scattering of electromagnetic waves in three dimensions.
• Karhunen-Loève approximation of random fields by generalized fast multipole methods
• A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems
• Analysis of the Hessian for inverse scattering problems: I. Inverse shape scattering of acoustic waves
• Analysis of the Hessian for inverse scattering problems: II. Inverse medium scattering of acoustic waves
• A Stochastic Newton MCMC Method for Large-Scale Statistical Inverse Problems with Application to Seismic Inversion
• Active Subspace Methods in Theory and Practice: Applications to Kriging Surfaces
• A Computational Framework for Infinite-Dimensional Bayesian Inverse Problems, Part II: Stochastic Newton MCMC with Application to Ice Sheet Flow Inverse Problems
• A-Optimal Design of Experiments for Infinite-Dimensional Bayesian Linear Inverse Problems with Regularized $\ell_0$-Sparsification
• Fast Algorithms for Bayesian Uncertainty Quantification in Large-Scale Linear Inverse Problems Based on Low-Rank Partial Hessian Approximations
• A Fast and Scalable Method for A-Optimal Design of Experiments for Infinite-dimensional Bayesian Nonlinear Inverse Problems
• Hessian-based model reduction for large-scale systems with initial-condition inputs
• Model Reduction for Large-Scale Systems with High-Dimensional Parametric Input Space
• A Probabilistic Subspace Bound with Application to Active Subspaces
• Deep learning in high dimension: Neural network expression rates for generalized polynomial chaos expansions in UQ
• Mean-Variance Risk-Averse Optimal Control of Systems Governed by PDEs with Random Parameter Fields Using Quadratic Approximations
• Reduced Basis Methods for Uncertainty Quantification
• The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
• Solving high-dimensional partial differential equations using deep learning
• hIPPYlib
• HESSIAN-BASED SAMPLING FOR HIGH-DIMENSIONAL MODEL REDUCTION
• Tensor Train Construction From Tensor Actions, With Application to Compression of Large High Order Derivative Tensors
• Gradient-Based Dimension Reduction of Multivariate Vector-Valued Functions
• Approximation of high-dimensional parametric PDEs
• A-optimal encoding weights for nonlinear inverse problems, with application to the Helmholtz inverse problem
• A Weighted Reduced Basis Method for Elliptic Partial Differential Equations with Random Input Data
• A Computational Framework for Infinite-Dimensional Bayesian Inverse Problems Part I: The Linearized Case, with Application to Global Seismic Inversion
• A Stochastic Collocation Method for Elliptic Partial Differential Equations with Random Input Data

This page was built for publication: Derivative-informed projected neural networks for high-dimensional parametric maps governed by PDEs