Neural-net-induced Gaussian process regression for function approximation and PDE solution
From MaRDI portal
Abstract: Neural-net-induced Gaussian process (NNGP) regression inherits both the high expressivity of deep neural networks (deep NNs) as well as the uncertainty quantification property of Gaussian processes (GPs). We generalize the current NNGP to first include a larger number of hyperparameters and subsequently train the model by maximum likelihood estimation. Unlike previous works on NNGP that targeted classification, here we apply the generalized NNGP to function approximation and to solving partial differential equations (PDEs). Specifically, we develop an analytical iteration formula to compute the covariance function of GP induced by deep NN with an error-function nonlinearity. We compare the performance of the generalized NNGP for function approximations and PDE solutions with those of GPs and fully-connected NNs. We observe that for smooth functions the generalized NNGP can yield the same order of accuracy with GP, while both NNGP and GP outperform deep NN. For non-smooth functions, the generalized NNGP is superior to GP and comparable or superior to deep NN.
Recommendations
- scientific article; zbMATH DE number 1304257
- DGM: a deep learning algorithm for solving partial differential equations
- Kernel-based approximation methods for partial differential equations: deterministic or stochastic problems?
- Numerical Gaussian processes for time-dependent and nonlinear partial differential equations
- Numerical solution of the parametric diffusion equation by deep neural networks
Cites work
- scientific article; zbMATH DE number 6378127 (Why is no real title available?)
- Bayesian learning for neural networks
- Computational investigations of low-discrepancy sequences
- Discovering variable fractional orders of advection-dispersion equations from field data using multi-fidelity Bayesian optimization
- Gaussian processes for machine learning (GPML) toolbox
- Gaussian processes for machine learning.
- Inferring solutions of differential equations using noisy multi-fidelity data
- Machine learning of linear differential equations using Gaussian processes
- Spectral and finite difference solutions of the Burgers equations
Cited in
(20)- Physics-informed neural networks for high-speed flows
- Physics-informed generative adversarial networks for stochastic differential equations
- Integrated Lagrangian and Eulerian 3D microstructure-explicit simulations for predicting macroscopic probabilistic SDT thresholds of energetic materials
- Parametric Gaussian process regression for big data
- Extreme learning machine collocation for the numerical solution of elliptic PDEs with sharp gradients
- Randomized Nyström approximation of non-negative self-adjoint operators
- Three ways to solve partial differential equations with neural networks — A review
- Symplectic neural networks in Taylor series form for Hamiltonian systems
- Discontinuity computing using physics-informed neural networks
- Interpretable machine learning: fundamental principles and 10 grand challenges
- Learning and meta-learning of stochastic advection–diffusion–reaction systems from sparse measurements
- Constraint-aware neural networks for Riemann problems
- The ADMM-PINNs algorithmic framework for nonsmooth PDE-constrained optimization: a deep learning approach
- B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
- Multi-fidelity Bayesian neural networks: algorithms and applications
- Monte Carlo fPINNs: deep learning method for forward and inverse problems involving high dimensional fractional partial differential equations
- Variational inference for nonlinear inverse problems via neural net kernels: comparison to Bayesian neural networks, application to topology optimization
- Deep learning of vortex-induced vibrations
- Learning elliptic partial differential equations with randomized linear algebra
- Fredholm neural networks
This page was built for publication: Neural-net-induced Gaussian process regression for function approximation and PDE solution
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2214653)
