Data driven governing equations approximation using deep neural networks
From MaRDI portal
Publication:2222362
DOI10.1016/J.JCP.2019.06.042zbMATH Open1455.65125arXiv1811.05537OpenAlexW2951629468WikidataQ127676911 ScholiaQ127676911MaRDI QIDQ2222362FDOQ2222362
Authors: Tong Qin, Kailiang Wu, Dongbin Xiu 
Publication date: 26 January 2021
Published in: Journal of Computational Physics (Search for Journal in Brave)
Abstract: We present a numerical framework for approximating unknown governing equations using observation data and deep neural networks (DNN). In particular, we propose to use residual network (ResNet) as the basic building block for equation approximation. We demonstrate that the ResNet block can be considered as a one-step method that is exact in temporal integration. We then present two multi-step methods, recurrent ResNet (RT-ResNet) method and recursive ReNet (RS-ResNet) method. The RT-ResNet is a multi-step method on uniform time steps, whereas the RS-ResNet is an adaptive multi-step method using variable time steps. All three methods presented here are based on integral form of the underlying dynamical system. As a result, they do not require time derivative data for equation recovery and can cope with relatively coarsely distributed trajectory data. Several numerical examples are presented to demonstrate the performance of the methods.
Full work available at URL: https://arxiv.org/abs/1811.05537
Recommendations
- Deep hidden physics models: deep learning of nonlinear partial differential equations
- PDE-Net 2.0: learning PDEs from data with a numeric-symbolic hybrid deep network
- Numerical aspects for approximating governing equations using data
- Learning data-driven discretizations for partial differential equations
- NeuralPDE: modelling dynamical systems from data
Artificial neural networks and deep learning (68T07) Time series analysis of dynamical systems (37M10) Dynamics induced by flows and semiflows (37C10) Numerical methods for ordinary differential equations (65L99)
Cites Work
- Detecting causality in complex ecosystems
- Title not available (Why is that?)
- Universal approximation bounds for superpositions of a sigmoidal function
- Deep learning
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Equation-free, coarse-grained multiscale computation: enabling microscopic simulators to perform system-level analysis
- Automated reverse engineering of nonlinear dynamical systems
- Numerical differentiation of noisy, nonsmooth data
- Title not available (Why is that?)
- Title not available (Why is that?)
- Nonlinear Laplacian spectral analysis for time series with intermittency and low-frequency variability
- Non-intrusive reduced order modeling of nonlinear problems using neural networks
- Hidden physics models: machine learning of nonlinear partial differential equations
- Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
- An artificial neural network as a troubled-cell indicator
- Machine learning of linear differential equations using Gaussian processes
- Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
- Deep hidden physics models: deep learning of nonlinear partial differential equations
- A machine learning approach for efficient uncertainty quantification using multiscale methods
- Neural networks and statistical learning
- Numerical aspects for approximating governing equations using data
- Model selection for dynamical systems via sparse regression and information criteria
- Learning partial differential equations via data discovery and sparse optimization
Cited In (80)
- Data-driven models of nonautonomous systems
- Learning nonparametric ordinary differential equations from noisy data
- Learning particle swarming models from data with Gaussian processes
- Quadrature rule based discovery of dynamics by data-driven denoising
- On the identification and optimization of nonsmooth superposition operators in semilinear elliptic PDEs
- A causality-DeepONet for causal responses of linear dynamical systems
- Data-driven synchronization-avoiding algorithms in the explicit distributed structural analysis of soft tissue
- SPADE4: sparsity and delay embedding based forecasting of epidemics
- Machine learning approaches for the solution of the Riemann problem in fluid dynamics: a case study
- DRIPS: a framework for dimension reduction and interpolation in parameter space
- Adaptive group Lasso neural network models for functions of few variables and time-dependent data
- Error analysis based on inverse modified differential equations for discovery of dynamics using linear multistep methods and deep learning
- A learned conservative semi-Lagrangian finite volume scheme for transport simulations
- Simulating partial differential equations with neural networks
- On the identifiability of nonlocal interaction kernels in first-order systems of interacting particles on Riemannian manifolds
- Learning the Dynamics for Unknown Hyperbolic Conservation Laws Using Deep Neural Networks
- On mathematical modeling in image reconstruction and beyond
- Machine learning methods for reduced order modeling
- Learning black- and gray-box chemotactic PDEs/closures from agent based Monte Carlo simulation data
- DNN modeling of partial differential equations with incomplete data
- A framework based on symbolic regression coupled with eXtended physics-informed neural networks for gray-box learning of equations of motion from data
- Data-driven modeling of partially observed biological systems
- Deep neural network based adaptive learning for switched systems
- A data-driven framework for learning hybrid dynamical systems
- Symbolic regression via neural networks
- Model-free inference of unseen attractors: reconstructing phase space features from a single noisy trajectory using Reservoir computing
- A Review of Data‐Driven Discovery for Dynamic Systems
- Structure-preserving recurrent neural networks for a class of Birkhoffian systems
- Robust topology optimization with low rank approximation using artificial neural networks
- Methods to recover unknown processes in partial differential equations using data
- Randomized Newton's method for solving differential equations based on the neural network discretization
- Weak SINDy for partial differential equations
- Discovering phase field models from image data with the pseudo-spectral physics informed neural networks
- Data-driven learning of nonautonomous systems
- A multi-scale DNN algorithm for nonlinear elliptic equations with multiple scales
- A Variational Neural Network Approach for Glacier Modelling with Nonlinear Rheology
- PDE-Net 2.0: learning PDEs from data with a numeric-symbolic hybrid deep network
- Robust modeling of unknown dynamical systems via ensemble averaged learning
- Deep learning of free boundary and Stefan problems
- Deep neural network modeling of unknown partial differential equations in nodal space
- The discovery of dynamics via linear multistep methods and deep learning: error estimation
- Data-driven sparse identification of nonlinear dynamical systems using linear multistep methods
- Deep-OSG: deep learning of operators in semigroup
- Using deep learning to extend the range of air pollution monitoring and forecasting
- On generalized residual network for deep learning of unknown dynamical systems
- Deep-learning based discovery of partial differential equations in integral form from sparse and noisy data
- Generalized Cell Mapping Method with Deep Learning for Global Analysis and Response Prediction of Dynamical Systems
- System identification through Lipschitz regularized deep neural networks
- Clustered active-subspace based local Gaussian process emulator for high-dimensional and complex computer models
- Dynamic calibration of differential equations using machine learning, with application to turbulence models
- A physics-informed operator regression framework for extracting data-driven continuum models
- Local parameter identification with neural ordinary differential equations
- Data-driven deep learning of partial differential equations in modal space
- A neural network multigrid solver for the Navier-Stokes equations
- Approximation properties of residual neural networks for Kolmogorov PDEs
- Structure-Preserving Method for Reconstructing Unknown Hamiltonian Systems From Trajectory Data
- Discovery of subdiffusion problem with noisy data via deep learning
- Numerical aspects for approximating governing equations using data
- A hybrid objective function for robustness of artificial neural networks -- estimation of parameters in a mechanical system
- Optimization with learning-informed differential equation constraints and its applications
- Sparse dynamical system identification with simultaneous structural parameters and initial condition estimation
- VPNets: volume-preserving neural networks for learning source-free dynamics
- Deep learning of biological models from data: applications to ODE models
- Discovery of dynamics using linear multistep methods
- Generative Adversarial Network for Probabilistic Forecast of Random Dynamical Systems
- Accuracy and architecture studies of residual neural network method for ordinary differential equations
- Learning and meta-learning of stochastic advection–diffusion–reaction systems from sparse measurements
- DeepParticle: learning invariant measure by a deep neural network minimizing Wasserstein distance on data generated from an interacting particle method
- Probabilistic solutions to DAEs learning from physical data
- Robust Identification of Differential Equations by Numerical Techniques from a Single Set of Noisy Observation
- Bayesian differential programming for robust systems identification under uncertainty
- PhyCRNet: physics-informed convolutional-recurrent network for solving spatiotemporal PDEs
- DL-PDE: Deep-Learning Based Data-Driven Discovery of Partial Differential Equations from Discrete and Noisy Data
- Simultaneous neural network approximation for smooth functions
- DLGA-PDE: discovery of PDEs with incomplete candidate library via combination of deep learning and genetic algorithm
- Galerkin neural networks: a framework for approximating variational equations with error control
- WeakIdent: weak formulation for identifying differential equation using narrow-fit and trimming
- Extended dynamic mode decomposition for inhomogeneous problems
- Artificial neural network approximations of Cauchy inverse problem for linear PDEs
- Learning data-driven discretizations for partial differential equations
Uses Software
This page was built for publication: Data driven governing equations approximation using deep neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2222362)
