Numerical Gaussian processes for time-dependent and nonlinear partial differential equations
From MaRDI portal
Publication:3130409
Abstract: We introduce the concept of numerical Gaussian processes, which we define as Gaussian processes with covariance functions resulting from temporal discretization of time-dependent partial differential equations. Numerical Gaussian processes, by construction, are designed to deal with cases where: (1) all we observe are noisy data on black-box initial conditions, and (2) we are interested in quantifying the uncertainty associated with such noisy data in our solutions to time-dependent partial differential equations. Our method circumvents the need for spatial discretization of the differential operators by proper placement of Gaussian process priors. This is an attempt to construct structured and data-efficient learning machines, which are explicitly informed by the underlying physics that possibly generated the observed data. The effectiveness of the proposed approach is demonstrated through several benchmark problems involving linear and nonlinear time-dependent operators. In all examples, we are able to recover accurate approximations of the latent solutions, and consistently propagate uncertainty, even in cases involving very long time integration.
Recommendations
- Bayesian numerical methods for nonlinear partial differential equations
- Hidden physics models: machine learning of nonlinear partial differential equations
- Deep hidden physics models: deep learning of nonlinear partial differential equations
- Inferring solutions of differential equations using noisy multi-fidelity data
- Gaussian functional regression for linear partial differential equations
Cites work
- scientific article; zbMATH DE number 4098703 (Why is no real title available?)
- scientific article; zbMATH DE number 3551792 (Why is no real title available?)
- scientific article; zbMATH DE number 1391397 (Why is no real title available?)
- scientific article; zbMATH DE number 5055767 (Why is no real title available?)
- scientific article; zbMATH DE number 3227378 (Why is no real title available?)
- 10.1162/15324430152748236
- A First Course in the Numerical Analysis of Differential Equations
- Brittleness of Bayesian inference under finite information in a continuous world
- Diagonally Implicit Runge–Kutta Methods for Stiff O.D.E.’s
- Gaussian processes for machine learning.
- Inferring solutions of differential equations using noisy multi-fidelity data
- Machine learning of linear differential equations using Gaussian processes
- Machine learning. A probabilistic perspective
- Machine learning: Trends, perspectives, and prospects
- Numerical Methods for Ordinary Differential Equations
- Posterior consistency for Gaussian process approximations of Bayesian posterior distributions
- Probabilistic numerics and uncertainty in computations
- Regularization algorithms for learning that are equivalent to multilayer networks
- Spectral and finite difference solutions of the Burgers equations
- Statistical analysis of differential equations: introducing probability measures on numerical solutions
- Theory of Reproducing Kernels
- Theory of reproducing kernels and applications
Cited in
(94)- Approximate Bayesian model inversion for PDEs with heterogeneous and state-dependent coefficients
- Deep hidden physics models: deep learning of nonlinear partial differential equations
- Neural-net-induced Gaussian process regression for function approximation and PDE solution
- Adversarial uncertainty quantification in physics-informed neural networks
- Methods to recover unknown processes in partial differential equations using data
- Physics-informed generative adversarial networks for stochastic differential equations
- Machine learning for a class of partial differential equations with multi-delays based on numerical Gaussian processes
- Autodifferentiable ensemble Kalman filters
- Generative stochastic modeling of strongly nonlinear flows with non-Gaussian statistics
- Augmented Gaussian random field: theory and computation
- On a multilevel Levenberg-Marquardt method for the training of artificial neural networks and its application to the solution of partial differential equations
- Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems
- Data-driven learning of nonautonomous systems
- A nonstationary space-time Gaussian process model for partially converged simulations
- Numerical solution and bifurcation analysis of nonlinear partial differential equations with extreme learning machines
- Numerical Approximations of the Allen-Cahn-Ohta-Kawasaki Equation with Modified Physics-Informed Neural Networks (Pinns)
- Three ways to solve partial differential equations with neural networks — A review
- The discovery of dynamics via linear multistep methods and deep learning: error estimation
- Deep learning of dynamics and signal-noise decomposition with time-stepping constraints
- Enhancing physics informed neural networks for solving Navier-Stokes equations
- scientific article; zbMATH DE number 4056689 (Why is no real title available?)
- Bi-Orthogonal fPINN: A Physics-Informed Neural Network Method for Solving Time-Dependent Stochastic Fractional PDEs
- Randomised one-step time integration methods for deterministic operator differential equations
- Hidden physics model for parameter estimation of elastic wave equations
- Solving second-order nonlinear evolution partial differential equations using deep learning
- Inferring solutions of differential equations using noisy multi-fidelity data
- Enforcing constraints for interpolation and extrapolation in generative adversarial networks
- Sparse identification of truncation errors
- Bayesian numerical methods for nonlinear partial differential equations
- \(\mathrm{U}^p\)-net: a generic deep learning-based time stepper for parameterized spatio-temporal dynamics
- Discovery of PDEs driven by data with sharp gradient or discontinuity
- An overview on deep learning-based approximation methods for partial differential equations
- Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
- Energy-informed graph transformer model for solid mechanical analyses
- Data-driven rogue waves and parameter discovery in the defocusing nonlinear Schrödinger equation with a potential using the PINN deep learning
- Data-driven identification of parametric partial differential equations
- System identification through Lipschitz regularized deep neural networks
- Normalizing field flows: solving forward and inverse stochastic differential equations using physics-informed flow models
- Revealing hidden dynamics from time-series data by ODENet
- Active training of physics-informed neural networks to aggregate and interpolate parametric solutions to the Navier-Stokes equations
- Sparse Gaussian processes for solving nonlinear PDEs
- Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems
- Learning biological dynamics from spatio-temporal data by Gaussian processes
- A Bayesian approach to modeling finite element discretization error
- Data-driven discovery of PDEs in complex datasets
- Numerical solution of nonlinear stochastic differential equations with fractional Brownian motion using fractional-order Genocchi deep neural networks
- Structure-Preserving Method for Reconstructing Unknown Hamiltonian Systems From Trajectory Data
- Gaussian process hydrodynamics
- Physics-informed cokriging: a Gaussian-process-regression-based multifidelity method for data-model convergence
- The deep learning Galerkin method for the general Stokes equations
- Adaptive activation functions accelerate convergence in deep and physics-informed neural networks
- Data-driven model identification using forcing-induced limit cycles
- Retracted: Model order reduction method based on machine learning for parameterized time-dependent partial differential equations
- A new class of high-order methods for fluid dynamics simulations using Gaussian process modeling: one-dimensional case
- Gaussian functional regression for linear partial differential equations
- Laplace-fPINNs: Laplace-based fractional physics-informed neural networks for solving forward and inverse problems of a time fractional equation
- Statistical Learning for Nonlinear Dynamical Systems with Applications to Aircraft-UAV Collisions
- Discovery of dynamics using linear multistep methods
- A hybrid MGA-MSGD ANN training approach for approximate solution of linear elliptic PDEs
- Wasserstein generative adversarial uncertainty quantification in physics-informed neural networks
- Interpretable machine learning: fundamental principles and 10 grand challenges
- Sparse polynomial chaos expansions using variational relevance vector machines
- Extracting Structured Dynamical Systems Using Sparse Optimization With Very Few Samples
- scientific article; zbMATH DE number 7370549 (Why is no real title available?)
- Covariance models and Gaussian process regression for the wave equation. Application to related inverse problems
- Stress-based topology optimization under uncertainty via simulation-based Gaussian process
- Learning and meta-learning of stochastic advection–diffusion–reaction systems from sparse measurements
- Leveraging viscous Hamilton-Jacobi PDEs for uncertainty quantification in scientific machine learning
- Parameter inference based on Gaussian processes informed by nonlinear partial differential equations
- A Kronecker product accelerated efficient sparse Gaussian process (E-SGP) for flow emulation
- Error analysis of kernel/GP methods for nonlinear and parametric PDEs
- The ADMM-PINNs algorithmic framework for nonsmooth PDE-constrained optimization: a deep learning approach
- The SPDE approach to Matérn fields: graph representations
- Optimally weighted loss functions for solving PDEs with neural networks
- Physics information aided kriging using stochastic simulation models
- A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems
- Machine Learning of Space-Fractional Differential Equations
- Images of Gaussian and other stochastic processes under closed, densely-defined, unbounded linear operators
- On the influence of over-parameterization in manifold based surrogates and deep neural operators
- Bayesian differential programming for robust systems identification under uncertainty
- SPINN: sparse, physics-based, and partially interpretable neural networks for PDEs
- A new artificial neural network method for solving Schrödinger equations on unbounded domains
- Learning sparse nonlinear dynamics via mixed-integer optimization
- Monte Carlo fPINNs: deep learning method for forward and inverse problems involving high dimensional fractional partial differential equations
- Deep learning of vortex-induced vibrations
- Bridging the gap: machine learning to resolve improperly modeled dynamics
- Scientific machine learning through physics-informed neural networks: where we are and what's next
- Bayesian learning of orthogonal embeddings for multi-fidelity Gaussian processes
- Bi-directional coupling between a PDE-domain and an adjacent data-domain equipped with multi-fidelity sensors
- Learning ``best kernels from data in Gaussian process regression. With application to aerodynamics
- Data-driven modeling of partially observed biological systems
- APIK: Active Physics-Informed Kriging Model with Partial Differential Equations
- Solving and learning nonlinear PDEs with Gaussian processes
- Using neural networks to accelerate the solution of the Boltzmann equation
This page was built for publication: Numerical Gaussian processes for time-dependent and nonlinear partial differential equations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3130409)
