Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
From MaRDI portal
Abstract: We are interested in the development of surrogate models for uncertainty quantification and propagation in problems governed by stochastic PDEs using a deep convolutional encoder-decoder network in a similar fashion to approaches considered in deep learning for image-to-image regression tasks. Since normal neural networks are data intensive and cannot provide predictive uncertainty, we propose a Bayesian approach to convolutional neural nets. A recently introduced variational gradient descent algorithm based on Stein's method is scaled to deep convolutional networks to perform approximate Bayesian inference on millions of uncertain network parameters. This approach achieves state of the art performance in terms of predictive accuracy and uncertainty quantification in comparison to other approaches in Bayesian neural networks as well as techniques that include Gaussian processes and ensemble methods even when the training data size is relatively small. To evaluate the performance of this approach, we consider standard uncertainty quantification benchmark problems including flow in heterogeneous media defined in terms of limited data-driven permeability realizations. The performance of the surrogate model developed is very good even though there is no underlying structure shared between the input (permeability) and output (flow/pressure) fields as is often the case in the image-to-image regression models used in computer vision problems. Studies are performed with an underlying stochastic input dimensionality up to where most other uncertainty quantification methods fail. Uncertainty propagation tasks are considered and the predictive output Bayesian statistics are compared to those obtained with Monte Carlo estimates.
Recommendations
- Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
- Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
- Conditional deep surrogate models for stochastic, high-dimensional, and multi-fidelity systems
- Scaling Up Bayesian Uncertainty Quantification for Inverse Problems Using Deep Neural Networks
- An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems
Cites work
- A Fast Learning Algorithm for Deep Belief Nets
- A machine learning approach for efficient uncertainty quantification using multiscale methods
- Bayesian model and dimension reduction for uncertainty propagation: applications in random media
- Cases for the nugget in modeling computer experiments
- Deep learning in fluid dynamics
- Gaussian processes for machine learning.
- Hyperband: a novel bandit-based approach to hyperparameter optimization
- Learning deep architectures for AI
- Multi-output local Gaussian process regression: applications to uncertainty quantification
- Multi-output separable Gaussian process: towards an efficient, fully Bayesian paradigm for uncertainty quantification
- Reducing the Dimensionality of Data with Neural Networks
- The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
- Visualizing data using t-SNE
Cited in
(only showing first 100 items - show all)- Machine learning in cardiovascular flows modeling: predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks
- Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
- Parametric generation of conditional geological realizations using generative neural networks
- Efficient uncertainty quantification for dynamic subsurface flow with surrogate by theory-guided neural network
- Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow
- An encoder-decoder deep surrogate for reverse time migration in seismic imaging under uncertainty
- Model reduction and neural networks for parametric PDEs
- Active learning Bayesian support vector regression model for global approximation
- Bayesian distillation of deep learning models
- PhyCRNet: physics-informed convolutional-recurrent network for solving spatiotemporal PDEs
- Mosaic flows: a transferable deep learning framework for solving PDEs on unseen domains
- A Bayesian multiscale deep learning framework for flows in random media
- A representative volume element network (RVE-net) for accelerating RVE analysis, microscale material identification, and defect characterization
- Learning phase field mean curvature flows with neural networks
- Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression
- An intelligent multi-fidelity surrogate-assisted multi-objective reservoir production optimization method based on transfer stacking
- Stochastic modeling of inhomogeneities in the aortic wall and uncertainty quantification using a Bayesian encoder-decoder surrogate
- Multifidelity data fusion in convolutional encoder/decoder networks
- Use of multifidelity training data and transfer learning for efficient construction of subsurface flow surrogate models
- A deep domain decomposition method based on Fourier features
- Data-driven soliton solutions and model parameters of nonlinear wave models via the conservation-law constrained neural network method
- GINNs: graph-informed neural networks for multiscale physics
- Solving inverse problems using conditional invertible neural networks
- An efficient neural network method with plane wave activation functions for solving Helmholtz equation
- Data-driven deep learning of partial differential equations in modal space
- A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables
- On generalized residual network for deep learning of unknown dynamical systems
- Transfer learning based multi-fidelity physics informed deep neural network
- Accelerated reactive transport simulations in heterogeneous porous media using Reaktoro and Firedrake
- Extended dynamic mode decomposition for inhomogeneous problems
- Solving and learning nonlinear PDEs with Gaussian processes
- Mesh-Conv: convolution operator with mesh resolution independence for flow field modeling
- Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
- Adaptive deep density approximation for Fokker-Planck equations
- Simulation of the 3D hyperelastic behavior of ventricular myocardium using a finite-element based neural-network approach
- Normalizing field flows: solving forward and inverse stochastic differential equations using physics-informed flow models
- A sample-efficient deep learning method for multivariate uncertainty qualification of acoustic-vibration interaction problems
- Surrogate and inverse modeling for two-phase flow in porous media via theory-guided convolutional neural network
- A heteroencoder architecture for prediction of failure locations in porous metals using variational inference
- Learning deep implicit Fourier neural operators (IFNOs) with applications to heterogeneous material modeling
- A data-driven surrogate to image-based flow simulations in porous media
- Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
- Surrogate modeling of high-dimensional problems via data-driven polynomial chaos expansions and sparse partial least square
- A bi-fidelity surrogate modeling approach for uncertainty propagation in three-dimensional hemodynamic simulations
- Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networks
- Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
- ConvPDE-UQ: convolutional neural networks with quantified uncertainty for heterogeneous elliptic partial differential equations on varied domains
- Data driven governing equations approximation using deep neural networks
- A physics-aware, probabilistic machine learning framework for coarse-graining high-dimensional systems in the small data regime
- Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems
- A mesh-free method for interface problems using the deep learning approach
- Reduced-order deep learning for flow dynamics. The interplay between deep learning and model reduction
- Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks
- Simulator-free solution of high-dimensional stochastic elliptic partial differential equations using deep neural networks
- Deep multiscale model learning
- Theory-guided auto-encoder for surrogate construction and inverse modeling
- Output-based adaptive aerodynamic simulations using convolutional neural networks
- Stein variational gradient descent with local approximations
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Conditional deep surrogate models for stochastic, high-dimensional, and multi-fidelity systems
- Deep learning of biological models from data: applications to ODE models
- A physics-informed variational DeepONet for predicting crack path in quasi-brittle materials
- Physics-informed Karhunen-Loéve and neural network approximations for solving inverse differential equation problems
- Wasserstein generative adversarial uncertainty quantification in physics-informed neural networks
- DeepParticle: learning invariant measure by a deep neural network minimizing Wasserstein distance on data generated from an interacting particle method
- A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems
- Scalable uncertainty quantification for deep operator networks using randomized priors
- LaSDI: parametric latent space dynamics identification
- The deep learning Galerkin method for the general Stokes equations
- Nonlocal kernel network (NKN): a stable and resolution-independent deep neural network
- A backward SDE method for uncertainty quantification in deep learning
- Surrogate modeling for Bayesian inverse problems based on physics-informed neural networks
- Deep reinforcement learning for optimal well control in subsurface systems with uncertain geology
- Stein's method meets computational statistics: a review of some recent developments
- opPINN: physics-informed neural network with operator learning to approximate solutions to the Fokker-Planck-Landau equation
- Forecasting of nonlinear dynamics based on symbolic invariance
- The Random Feature Model for Input-Output Maps between Banach Spaces
- Deep learning of vortex-induced vibrations
- Variational inference formulation for a model-free simulation of a dynamical system with unknown parameters by a recurrent neural network
- What machine learning can do for computational solid mechanics
- A deep learning approach to Reduced Order Modelling of parameter dependent partial differential equations
- slimTrain---A Stochastic Approximation Method for Training Separable Deep Neural Networks
- Deep neural network surrogates for nonsmooth quantities of interest in shape uncertainty quantification
- Physics-Informed Probabilistic Learning of Linear Embeddings of Nonlinear Dynamics with Guaranteed Stability
- Discovery of dynamics using linear multistep methods
- GAN-based priors for quantifying uncertainty in supervised learning
- An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems
- Cholesky-based experimental design for Gaussian process and kernel-based emulation and calibration
- Physics-informed generative adversarial networks for stochastic differential equations
- Bayesian model and dimension reduction for uncertainty propagation: applications in random media
- VI-DGP: a variational inference method with deep generative prior for solving high-dimensional inverse problems
- Epistemic uncertainty quantification in deep learning classification by the delta method
- Transformers for modeling physical systems
- Conditional variational autoencoder with Gaussian process regression recognition for parametric models
- Probabilistic partition of unity networks for high‐dimensional regression problems
- Adaptive experimental design for multi‐fidelity surrogate modeling of multi‐disciplinary systems
- Deep learning methods for partial differential equations and related parameter identification problems
- Deep capsule encoder–decoder network for surrogate modeling and uncertainty quantification
- Multifidelity approaches for uncertainty quantification
- Physics-informed data-driven model for fluid flow in porous media
Describes a project that uses
Uses Software
This page was built for publication: Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1721865)
