Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
From MaRDI portal
Publication:2002273
Abstract: State-of-the-art computer codes for simulating real physical systems are often characterized by a vast number of input parameters. Performing uncertainty quantification (UQ) tasks with Monte Carlo (MC) methods is almost always infeasible because of the need to perform hundreds of thousands or even millions of forward model evaluations in order to obtain convergent statistics. One, thus, tries to construct a cheap-to-evaluate surrogate model to replace the forward model solver. For systems with large numbers of input parameters, one has to deal with the curse of dimensionality - the exponential increase in the volume of the input space, as the number of parameters increases linearly. In this work, we demonstrate the use of deep neural networks (DNN) to construct surrogate models for numerical simulators. We parameterize the structure of the DNN in a manner that lends the DNN surrogate the interpretation of recovering a low dimensional nonlinear manifold. The model response is a parameterized nonlinear function of the low dimensional projections of the input. We think of this low dimensional manifold as a nonlinear generalization of the notion of the active subspace. Our approach is demonstrated with a problem on uncertainty propagation in a stochastic elliptic partial differential equation (SPDE) with uncertain diffusion coefficient. We deviate from traditional formulations of the SPDE problem by not imposing a specific covariance structure on the random diffusion coefficient. Instead, we attempt to solve a more challenging problem of learning a map between an arbitrary snapshot of the diffusion field and the response.
Recommendations
- Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
- Deep neural network surrogates for nonsmooth quantities of interest in shape uncertainty quantification
- Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
- Deep learning in high dimension: neural network expression rates for generalized polynomial chaos expansions in UQ
- An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems
Cites work
- scientific article; zbMATH DE number 6378127 (Why is no real title available?)
- scientific article; zbMATH DE number 1191619 (Why is no real title available?)
- scientific article; zbMATH DE number 1304256 (Why is no real title available?)
- scientific article; zbMATH DE number 2117879 (Why is no real title available?)
- Active subspace methods in theory and practice: applications to kriging surfaces
- Adaptive subgradient methods for online learning and stochastic optimization
- AlexNet
- Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
- Deep learning
- Discovering an active subspace in a single‐diode solar cell model
- Exploiting active subspaces to quantify uncertainty in the numerical simulation of the hyshot II scramjet
- Gaussian processes for machine learning.
- Gaussian processes with built-in dimensionality reduction: applications to high-dimensional uncertainty propagation
- Introduction to uncertainty quantification
- Inverse regression-based uncertainty quantification algorithms for high-dimensional models: theory and practice
- Kernel principal component analysis for stochastic input model generation
- Large-scale machine learning with stochastic gradient descent
- Modeling uncertainty in flow simulations via generalized polynomial chaos.
- Modeling uncertainty in steady state diffusion problems via generalized polynomial chaos
- Monte Carlo analysis of inverse problems
- Monte Carlo strategies in scientific computing.
- Multi-output local Gaussian process regression: applications to uncertainty quantification
- Multi-output separable Gaussian process: towards an efficient, fully Bayesian paradigm for uncertainty quantification
- Multidimensional adaptive relevance vector machines for uncertainty quantification
- Multilayer feedforward networks are universal approximators
- Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling
- Nonlinear methods of spectral analysis. 2nd corrected and updated edition
- Predicting the output from a complex computer code when fast approximations are available
- Principal component analysis.
- Regularization and Variable Selection Via the Elastic Net
- Survey of multifidelity methods in uncertainty propagation, inference, and optimization
- The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
- Uncertainty Quantification and Polynomial Chaos Techniques in Computational Fluid Dynamics
- Uncertainty propagation using infinite mixture of gaussian processes and variational Bayesian inference
Cited in
(only showing first 100 items - show all)- scientific article; zbMATH DE number 7326170 (Why is no real title available?)
- Deep learning of vortex-induced vibrations
- ConvPDE-UQ: convolutional neural networks with quantified uncertainty for heterogeneous elliptic partial differential equations on varied domains
- Bayesian neural networks for predicting uncertainty in full-field material response
- A review of recent advances in surrogate models for uncertainty quantification of high-dimensional engineering applications
- Uncertainty quantification for noisy inputs-outputs in physics-informed neural networks and neural operators
- Solving inverse problems using conditional invertible neural networks
- From obstacle problems to neural insights: feedforward neural network modeling of ice thickness
- Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
- An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems
- A deep learning driven pseudospectral PCE based FFT homogenization algorithm for complex microstructures
- Bayesian learning of orthogonal embeddings for multi-fidelity Gaussian processes
- Data-driven solutions and parameter discovery of the Sasa-Satsuma equation via the physics-informed neural networks method
- Data-driven modeling of partially observed biological systems
- Extended dynamic mode decomposition for inhomogeneous problems
- Variational inference with NoFAS: normalizing flow with adaptive surrogate for computationally expensive models
- Conditional Karhunen-Loève regression model with basis adaptation for high-dimensional problems: uncertainty quantification and inverse modeling
- Addressing discontinuous root-finding for subsequent differentiability in machine learning, inverse problems, and control
- Numerical weighted integration of functions having mixed smoothness
- Stochastic modeling of inhomogeneities in the aortic wall and uncertainty quantification using a Bayesian encoder-decoder surrogate
- Multi-Fidelity Uncertainty Propagation Approach for Multi-Dimensional Correlated Flow Field Responses
- Simulation of the 3D hyperelastic behavior of ventricular myocardium using a finite-element based neural-network approach
- Deep capsule encoder–decoder network for surrogate modeling and uncertainty quantification
- A Model-Constrained Tangent Slope Learning Approach for Dynamical Systems
- Deep learning in high dimension: neural network expression rates for generalized polynomial chaos expansions in UQ
- Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
- Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
- Bayesian model and dimension reduction for uncertainty propagation: applications in random media
- Stacking Designs: Designing Multifidelity Computer Experiments with Target Predictive Accuracy
- Deep convolutional Ritz method: parametric PDE surrogates without labeled data
- A framework for strategic discovery of credible neural network surrogate models under uncertainty
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- VI-DGP: a variational inference method with deep generative prior for solving high-dimensional inverse problems
- Dimensionality reduction for complex models via Bayesian compressive sensing
- Probabilistic deep learning for real-time large deformation simulations
- Learning nonlocal constitutive models with neural networks
- Multifidelity multilevel Monte Carlo to accelerate approximate Bayesian parameter inference for partially observed stochastic processes
- Scalable uncertainty quantification for deep operator networks using randomized priors
- slimTrain---A Stochastic Approximation Method for Training Separable Deep Neural Networks
- Uncertainty quantification in scientific machine learning: methods, metrics, and comparisons
- Solving the discretised neutron diffusion equations using neural networks
- Learning to solve Bayesian inverse problems: an amortized variational inference approach using Gaussian and flow guides
- Wave-packet behaviors of the defocusing nonlinear Schrödinger equation based on the modified physics-informed neural networks
- Train Like a (Var)Pro: Efficient Training of Neural Networks with Variable Projection
- TDOR-MPINNs: multi-output physics-informed neural networks based on time differential order reduction for solving coupled Klein-Gordon-Zakharov systems
- Deep learning of free boundary and Stefan problems
- Random field of homogeneous and multi-material structures by the smoothed finite element method and Karhunen-Loève expansion
- LaSDI: parametric latent space dynamics identification
- A bi-fidelity surrogate modeling approach for uncertainty propagation in three-dimensional hemodynamic simulations
- Collocation approximation by deep neural ReLU networks for parametric and stochastic PDEs with lognormal inputs
- Quality measures for the evaluation of machine learning architectures on the quantification of epistemic and aleatoric uncertainties in complex dynamical systems
- The Random Feature Model for Input-Output Maps between Banach Spaces
- Emulation of CPU-demanding reactive transport models: a comparison of Gaussian processes, polynomial chaos expansion, and deep neural networks
- GINNs: graph-informed neural networks for multiscale physics
- On generalized residual network for deep learning of unknown dynamical systems
- Operator learning using random features: a tool for scientific computing
- NeuralUQ: A Comprehensive Library for Uncertainty Quantification in Neural Differential Equations and Operators
- TNet: A Model-Constrained Tikhonov Network Approach for Inverse Problems
- Bayesian Deep Learning Framework for Uncertainty Quantification in Stochastic Partial Differential Equations
- A physics-aware, probabilistic machine learning framework for coarse-graining high-dimensional systems in the small data regime
- Deep weighted Monte Carlo: a hybrid option pricing framework using neural networks
- Scalable conditional deep inverse Rosenblatt transports using tensor trains and gradient-based dimension reduction
- Level Set Learning with Pseudoreversible Neural Networks for Nonlinear Dimension Reduction in Function Approximation
- Image inversion and uncertainty quantification for constitutive laws of pattern formation
- A statistician teaches deep learning
- Bayesian reduced-order deep learning surrogate model for dynamic systems described by partial differential equations
- Numerical solution of the parametric diffusion equation by deep neural networks
- Simulator-free solution of high-dimensional stochastic elliptic partial differential equations using deep neural networks
- Mutual information for explainable deep learning of multiscale systems
- Stein variational gradient descent with local approximations
- PPINN: parareal physics-informed neural network for time-dependent PDEs
- Efficient uncertainty quantification for dynamic subsurface flow with surrogate by theory-guided neural network
- Physics-informed machine learning models for predicting the progress of reactive-mixing
- Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
- Data-driven deep learning of partial differential equations in modal space
- Neural and spectral operator surrogates: unified construction and expression rate bounds
- Parametric dynamic mode decomposition for reduced order modeling
- Bi-fidelity modeling of uncertain and partially unknown systems using DeepONets
- Surrogate and inverse modeling for two-phase flow in porous media via theory-guided convolutional neural network
- ReLU neural network Galerkin BEM
- A sample-efficient deep learning method for multivariate uncertainty qualification of acoustic-vibration interaction problems
- Surrogate modeling for Bayesian inverse problems based on physics-informed neural networks
- Data driven governing equations approximation using deep neural networks
- Transformers for modeling physical systems
- Surrogate modeling for high dimensional uncertainty propagation via deep kernel polynomial chaos expansion
- Deep ReLU neural network approximation in Bochner spaces and applications to parametric PDEs
- Probabilistic partition of unity networks for high‐dimensional regression problems
- Conditional deep surrogate models for stochastic, high-dimensional, and multi-fidelity systems
- Estimation of an improved surrogate model in uncertainty quantification by neural networks
- Deep neural network surrogates for nonsmooth quantities of interest in shape uncertainty quantification
- Deep Importance Sampling Using Tensor Trains with Application to a Priori and a Posteriori Rare Events
- A higher-order singular value decomposition tensor emulator for spatiotemporal simulators
- Deep learning of biological models from data: applications to ODE models
- Dual order-reduced Gaussian process emulators (DORGP) for quantifying high-dimensional uncertain crack growth using limited and noisy data
- Functional PCA and deep neural networks-based Bayesian inverse uncertainty quantification with transient experimental data
- Quantification on the generalization performance of deep neural network with Tychonoff separation axioms
- An hp‐adaptive multi‐element stochastic collocation method for surrogate modeling with information re‐use
- Active learning Bayesian support vector regression model for global approximation
- Bi-fidelity variational auto-encoder for uncertainty quantification
- Polynomial chaos expansions on principal geodesic Grassmannian submanifolds for surrogate modeling and uncertainty quantification
Describes a project that uses
Uses Software
This page was built for publication: Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2002273)
