Deep Neural Network Surrogates for Nonsmooth Quantities of Interest in Shape Uncertainty Quantification

DOI10.1137/21M1393078zbMath1496.35467arXiv2101.07023OpenAlexW3126137513MaRDI QIDQ5097855

Laura Scarabosio

Publication date: 1 September 2022

Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2101.07023

zbMATH Keywords

Helmholtzmachine learningrandom interfacedeep neural networks

Mathematics Subject Classification ID

Probabilistic methods, particle methods, etc. for boundary value problems involving PDEs (65N75) Boundary value problems for second-order elliptic equations (35J25) PDEs with randomness, stochastic partial differential equations (35R60) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Approximation algorithms (68W25)

Uses Software

• FEniCS
• PyTorch
• Adam
• DeepXDE

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Analysis of the domain mapping method for elliptic diffusion problems on random domains
• Characterization of discontinuities in high-dimensional stochastic problems on adaptive sparse grids
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• Automated solution of differential equations by the finite element method. The FEniCS book
• Deep learning observables in computational fluid dynamics
• A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs
• Heaviside enriched extended stochastic FEM for problems with uncertain material interfaces
• Sparse second moment analysis for elliptic problems in stochastic domains
• The multi-element probabilistic collocation method (ME-PCM): Error analysis and applications
• An adaptive hierarchical sparse grid collocation algorithm for the solution of stochastic differential equations
• An \(H^S\)-regularity result for the gradient of solutions to elliptic equations with mixed boundary conditions
• Large deformation shape uncertainty quantification in acoustic scattering
• Non-intrusive reduced order modeling of nonlinear problems using neural networks
• Inferring solutions of differential equations using noisy multi-fidelity data
• Multilayer feedforward networks are universal approximators
• The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
• Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
• Analytic regularity and collocation approximation for elliptic PDEs with random domain deformations
• A stochastic collocation approach for parabolic PDEs with random domain deformations
• Numerical solution of the parametric diffusion equation by deep neural networks
• Proof that deep artificial neural networks overcome the curse of dimensionality in the numerical approximation of Kolmogorov partial differential equations with constant diffusion and nonlinear drift coefficients
• A theoretical analysis of deep neural networks and parametric PDEs
• Nonlinear approximation and (deep) ReLU networks
• Least-squares Padé approximation of parametric and stochastic Helmholtz maps
• Optimal approximation of piecewise smooth functions using deep ReLU neural networks
• Data-driven discovery of PDEs in complex datasets
• A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
• Uncertainty quantification of discontinuous outputs via a non-intrusive bifidelity strategy
• Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders
• Regularity of the Maxwell equations in heterogeneous media and Lipschitz domains
• Uncertainty quantification of geochemical and mechanical compaction in layered sedimentary basins
• An energy approach to the solution of partial differential equations in computational mechanics via machine learning: concepts, implementation and applications
• Bayesian inference of earthquake parameters from buoy data using a polynomial chaos-based surrogate
• An artificial neural network as a troubled-cell indicator
• A fictitious domain approach to the numerical solution of PDEs in stochastic domains
• An adaptive multi-element generalized polynomial chaos method for stochastic differential equations
• Stochastic analysis of transport in tubes with rough walls
• An extended stochastic finite element method for solving stochastic partial differential equations on random domains
• Uncertainty Quantification given Discontinuous Model Response and a Limited Number of Model Runs
• X-SFEM, a computational technique based on X-FEM to deal with random shapes
• On the Expressive Power of Deep Architectures
• Hyperspherical Sparse Approximation Techniques for High-Dimensional Discontinuity Detection
• Numerical Methods for Differential Equations in Random Domains
• Elliptic Partial Differential Equations of Second Order
• The Perfectly Matched Layer in Curvilinear Coordinates
• Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
• Electromagnetic wave scattering by random surfaces: Shape holomorphy
• Shape Holomorphy of the Stationary Navier--Stokes Equations
• Convergence analysis of Padé approximations for Helmholtz frequency response problems
• Deep learning in high dimension: Neural network expression rates for generalized polynomial chaos expansions in UQ
• Acoustic transmission problems: Wavenumber-explicit bounds and resonance-free regions
• Adaptive Anisotropic Spectral Stochastic Methods for Uncertain Scalar Conservation Laws
• The Gap between Theory and Practice in Function Approximation with Deep Neural Networks
• Enhancing Accuracy of Deep Learning Algorithms by Training with Low-Discrepancy Sequences
• Analysis of the Generalization Error: Empirical Risk Minimization over Deep Artificial Neural Networks Overcomes the Curse of Dimensionality in the Numerical Approximation of Black--Scholes Partial Differential Equations
• VPS: VORONOI PIECEWISE SURROGATE MODELS FOR HIGH-DIMENSIONAL DATA FITTING
• MULTILEVEL MONTE CARLO ON A HIGH-DIMENSIONAL PARAMETER SPACE FOR TRANSMISSION PROBLEMS WITH GEOMETRIC UNCERTAINTIES
• Deep neural network expression of posterior expectations in Bayesian PDE inversion
• DeepXDE: A Deep Learning Library for Solving Differential Equations
• Deep Network Approximation for Smooth Functions
• First order $k$-th moment finite element analysis of nonlinear operator equations with stochastic data
• First order second moment analysis for stochastic interface problems based on low-rank approximation
• Approximation by superpositions of a sigmoidal function
• Acoustic and electromagnetic equations. Integral representations for harmonic problems

This page was built for publication: Deep Neural Network Surrogates for Nonsmooth Quantities of Interest in Shape Uncertainty Quantification