Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
From MaRDI portal
Publication:4558167
zbMath1439.68021arXiv1801.06637MaRDI QIDQ4558167
Publication date: 21 November 2018
Abstract: A long-standing problem at the interface of artificial intelligence and applied mathematics is to devise an algorithm capable of achieving human level or even superhuman proficiency in transforming observed data into predictive mathematical models of the physical world. In the current era of abundance of data and advanced machine learning capabilities, the natural question arises: How can we automatically uncover the underlying laws of physics from high-dimensional data generated from experiments? In this work, we put forth a deep learning approach for discovering nonlinear partial differential equations from scattered and potentially noisy observations in space and time. Specifically, we approximate the unknown solution as well as the nonlinear dynamics by two deep neural networks. The first network acts as a prior on the unknown solution and essentially enables us to avoid numerical differentiations which are inherently ill-conditioned and unstable. The second network represents the nonlinear dynamics and helps us distill the mechanisms that govern the evolution of a given spatiotemporal data-set. We test the effectiveness of our approach for several benchmark problems spanning a number of scientific domains and demonstrate how the proposed framework can help us accurately learn the underlying dynamics and forecast future states of the system. In particular, we study the Burgers', Korteweg-de Vries (KdV), Kuramoto-Sivashinsky, nonlinear Schr"{o}dinger, and Navier-Stokes equations.
Full work available at URL: https://arxiv.org/abs/1801.06637
nonlinear dynamicsbig datapredictive modelingsystems identificationdata-driven scientific discoveryphysics informed machine learning
Artificial neural networks and deep learning (68T07) KdV equations (Korteweg-de Vries equations) (35Q53) Navier-Stokes equations (35Q30) NLS equations (nonlinear Schrödinger equations) (35Q55) Nonlinear higher-order PDEs (35G20) Computational aspects of data analysis and big data (68T09)
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Detecting Causality in Complex Ecosystems
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- A fast immersed boundary method using a nullspace approach and multi-domain far-field boundary conditions
- Spectral and finite difference solutions of the Burgers equations
- Inferring solutions of differential equations using noisy multi-fidelity data
- Machine learning of linear differential equations using Gaussian processes
- Hidden physics models: machine learning of nonlinear partial differential equations
- The immersed boundary method: a projection approach
- Spectral properties of dynamical systems, model reduction and decompositions
- Equation-free, coarse-grained multiscale computation: enabling microscopic simulators to perform system-level analysis
- Compressive Sensing and Low-Rank Libraries for Classification of Bifurcation Regimes in Nonlinear Dynamical Systems
- Applied Koopmanism
- Normal forms for reduced stochastic climate models
- Numerical Gaussian Processes for Time-Dependent and Nonlinear Partial Differential Equations
- Automated reverse engineering of nonlinear dynamical systems
- Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
- Nonlinear Laplacian spectral analysis for time series with intermittency and low-frequency variability
- Analysis of Fluid Flows via Spectral Properties of the Koopman Operator
- Sparse dynamics for partial differential equations
- Compressed modes for variational problems in mathematics and physics
- On the Compressive Spectral Method
Related Items (only showing first 100 items - show all)
Sparse Deep Neural Network for Nonlinear Partial Differential Equations ⋮ Learning theory for inferring interaction kernels in second-order interacting agent systems ⋮ Physics-incorporated convolutional recurrent neural networks for source identification and forecasting of dynamical systems ⋮ Three ways to solve partial differential equations with neural networks — A review ⋮ Deep learning methods for partial differential equations and related parameter identification problems ⋮ On the use of graph neural networks and shape‐function‐based gradient computation in the deep energy method ⋮ CD-ROM: complemented deep -- reduced order model ⋮ Deep-OSG: deep learning of operators in semigroup ⋮ DNN modeling of partial differential equations with incomplete data ⋮ Derivative-based SINDy (DSINDy): addressing the challenge of discovering governing equations from noisy data ⋮ Principled interpolation of Green's functions learned from data ⋮ The robust physics-informed neural networks for a typical fourth-order phase field model ⋮ Physics-informed deep learning for melting heat transfer analysis with model-based transfer learning ⋮ Pre-training strategy for solving evolution equations based on physics-informed neural networks ⋮ Deep learning discrete calculus (DLDC): a family of discrete numerical methods by universal approximation for STEM education to frontier research ⋮ Respecting causality for training physics-informed neural networks ⋮ A novel physics-informed deep learning strategy with local time-updating discrete scheme for multi-dimensional forward and inverse consolidation problems ⋮ Koopman operator learning using invertible neural networks ⋮ A Review of Data‐Driven Discovery for Dynamic Systems ⋮ \(\mathrm{U}^p\)-net: a generic deep learning-based time stepper for parameterized spatio-temporal dynamics ⋮ Discovery of PDEs driven by data with sharp gradient or discontinuity ⋮ Physics-informed deep learning for three-dimensional transient heat transfer analysis of functionally graded materials ⋮ Koopman Operator Inspired Nonlinear System Identification ⋮ Deep Weak Approximation of SDEs: A Spatial Approximation Scheme for Solving Kolmogorov Equations ⋮ Evolutionary sparse data-driven discovery of multibody system dynamics ⋮ Bayesian deep learning for partial differential equation parameter discovery with sparse and noisy data ⋮ An Adaptive Physics-Informed Neural Network with Two-Stage Learning Strategy to Solve Partial Differential Equations ⋮ Learning the Dynamics for Unknown Hyperbolic Conservation Laws Using Deep Neural Networks ⋮ Physics-informed ConvNet: learning physical field from a shallow neural network ⋮ Unnamed Item ⋮ Data-Driven Identification of Parametric Partial Differential Equations ⋮ Learning particle swarming models from data with Gaussian processes ⋮ Long-time prediction of nonlinear parametrized dynamical systems by deep learning-based reduced order models ⋮ Koopman neural operator as a mesh-free solver of non-linear partial differential equations ⋮ PDE-READ: human-readable partial differential equation discovery using deep learning ⋮ Preserving bifurcations through moment closures ⋮ \texttt{Weak-PDE-LEARN}: a weak form based approach to discovering PDEs from noisy, limited data ⋮ Data-driven modeling of partially observed biological systems ⋮ PDNNs: the parallel deep neural networks for the Navier-Stokes equations coupled with heat equation ⋮ Deep neural network solutions for oscillatory Fredholm integral equations ⋮ Data-driven models of nonautonomous systems ⋮ Enhancing physics informed neural networks for solving Navier-Stokes equations ⋮ Meta-auto-decoder: a meta-learning-based reduced order model for solving parametric partial differential equations ⋮ Graph neural reaction diffusion models ⋮ Uncertainty quantification of graph convolution neural network models of evolving processes ⋮ Statistical Learning for Nonlinear Dynamical Systems with Applications to Aircraft-UAV Collisions ⋮ RandONets: shallow networks with random projections for learning linear and nonlinear operators ⋮ Simple yet effective adaptive activation functions for physics-informed neural networks ⋮ Slow invariant manifolds of fast-slow systems of ODEs with physics-informed neural networks ⋮ A deep learning method for incompressible fluid equations and the coupling problems ⋮ Learning in Modal Space: Solving Time-Dependent Stochastic PDEs Using Physics-Informed Neural Networks ⋮ Machine Learning of Space-Fractional Differential Equations ⋮ Multilevel Fine-Tuning: Closing Generalization Gaps in Approximation of Solution Maps under a Limited Budget for Training Data ⋮ Solving forward and inverse problems of the logarithmic nonlinear Schrödinger equation with \(\mathcal{PT}\)-symmetric harmonic potential via deep learning ⋮ Data-driven deep learning of partial differential equations in modal space ⋮ On some neural network architectures that can represent viscosity solutions of certain high dimensional Hamilton-Jacobi partial differential equations ⋮ On generalized residual network for deep learning of unknown dynamical systems ⋮ A derivative-free method for solving elliptic partial differential equations with deep neural networks ⋮ Physics-informed machine learning with conditional Karhunen-Loève expansions ⋮ DeepMoD: deep learning for model discovery in noisy data ⋮ A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs ⋮ On semi analytical and numerical simulations for a mathematical biological model; the time-fractional nonlinear Kolmogorov-Petrovskii-Piskunov (KPP) equation ⋮ System identification through Lipschitz regularized deep neural networks ⋮ Hybrid FEM-NN models: combining artificial neural networks with the finite element method ⋮ MIM: a deep mixed residual method for solving high-order partial differential equations ⋮ DeLISA: deep learning based iteration scheme approximation for solving PDEs ⋮ A two-stage physics-informed neural network method based on conserved quantities and applications in localized wave solutions ⋮ Thermodynamically consistent physics-informed neural networks for hyperbolic systems ⋮ When and why PINNs fail to train: a neural tangent kernel perspective ⋮ Deep neural network modeling of unknown partial differential equations in nodal space ⋮ Revealing hidden dynamics from time-series data by ODENet ⋮ CAN-PINN: a fast physics-informed neural network based on coupled-automatic-numerical differentiation method ⋮ Discretization of parameter identification in PDEs using neural networks ⋮ Choice of an initial guess for Newton's method to solve nonlinear differential equations ⋮ Data-driven discovery of governing equations for transient heat transfer analysis ⋮ Discovery of subdiffusion problem with noisy data via deep learning ⋮ Numerical bifurcation analysis of PDEs from lattice Boltzmann model simulations: a parsimonious machine learning approach ⋮ Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms ⋮ Solving elliptic equations with Brownian motion: bias reduction and temporal difference learning ⋮ Blood and breath alcohol concentration from transdermal alcohol biosensor data: estimation and uncertainty quantification via forward and inverse filtering for a covariate-dependent, physics-informed, hidden Markov model* ⋮ Scientific machine learning through physics-informed neural networks: where we are and what's next ⋮ Modelling spatiotemporal dynamics from Earth observation data with neural differential equations ⋮ Five semi analytical and numerical simulations for the fractional nonlinear space-time telegraph equation ⋮ Fractional Chebyshev deep neural network (FCDNN) for solving differential models ⋮ A data-driven surrogate to image-based flow simulations in porous media ⋮ Discovering phase field models from image data with the pseudo-spectral physics informed neural networks ⋮ An unsupervised deep learning approach to solving partial integro-differential equations ⋮ A hybrid inference system for improved curvature estimation in the level-set method using machine learning ⋮ Physics and equality constrained artificial neural networks: application to forward and inverse problems with multi-fidelity data fusion ⋮ Operator inference and physics-informed learning of low-dimensional models for incompressible flows ⋮ The Discovery of Dynamics via Linear Multistep Methods and Deep Learning: Error Estimation ⋮ ModalPINN: an extension of physics-informed neural networks with enforced truncated Fourier decomposition for periodic flow reconstruction using a limited number of imperfect sensors ⋮ Deep Neural Network Surrogates for Nonsmooth Quantities of Interest in Shape Uncertainty Quantification ⋮ A physically constrained variational autoencoder for geochemical pattern recognition ⋮ Wavelet neural operator for solving parametric partial differential equations in computational mechanics problems ⋮ Sparse dynamical system identification with simultaneous structural parameters and initial condition estimation ⋮ Extension of dynamic mode decomposition for dynamic systems with incomplete information based on t-model of optimal prediction ⋮ CPINNs: a coupled physics-informed neural networks for the closed-loop geothermal system ⋮ Neural network architectures using min-plus algebra for solving certain high-dimensional optimal control problems and Hamilton-Jacobi PDEs ⋮ Unnamed Item
Uses Software
This page was built for publication: Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
