Deep learning of dynamics and signal-noise decomposition with time-stepping constraints

DOI10.1016/j.jcp.2019.06.056zbMath1452.68170arXiv1808.02578OpenAlexW2887383111WikidataQ127565916 ScholiaQ127565916MaRDI QIDQ2222431

Steven L. Brunton, Samuel H. Rudy, J. Nathan Kutz

Publication date: 27 January 2021

Published in: Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1808.02578

zbMATH Keywords

system identificationdynamical systemsneural networksmachine learningdeep learningdata-driven models

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Time series analysis of dynamical systems (37M10) Qualitative investigation and simulation of ordinary differential equation models (34C60) Complex behavior and chaotic systems of ordinary differential equations (34C28)

Related Items (40)

Data-driven deep learning of partial differential equations in modal space ⋮ On generalized residual network for deep learning of unknown dynamical systems ⋮ Optimal Transport for Parameter Identification of Chaotic Dynamics via Invariant Measures ⋮ SubTSBR to tackle high noise and outliers for data-driven discovery of differential equations ⋮ Weak SINDy for partial differential equations ⋮ Extended dynamic mode decomposition for inhomogeneous problems ⋮ Deep neural network modeling of unknown partial differential equations in nodal space ⋮ Revealing hidden dynamics from time-series data by ODENet ⋮ Discovery of subdiffusion problem with noisy data via deep learning ⋮ A hybrid objective function for robustness of artificial neural networks -- estimation of parameters in a mechanical system ⋮ The Discovery of Dynamics via Linear Multistep Methods and Deep Learning: Error Estimation ⋮ DeepParticle: learning invariant measure by a deep neural network minimizing Wasserstein distance on data generated from an interacting particle method ⋮ Data-driven sparse identification of nonlinear dynamical systems using linear multistep methods ⋮ Bandwidth selection for kernel density estimation of fat-tailed and skewed distributions ⋮ Discovering interpretable Lagrangian of dynamical systems from data ⋮ DNN modeling of partial differential equations with incomplete data ⋮ Accuracy and architecture studies of residual neural network method for ordinary differential equations ⋮ Simultaneous Identification and Denoising of Dynamical Systems ⋮ Active Operator Inference for Learning Low-Dimensional Dynamical-System Models from Noisy Data ⋮ Neural Galerkin schemes with active learning for high-dimensional evolution equations ⋮ Kernel functions embed into the autoencoder to identify the sparse models of nonlinear dynamics ⋮ Unnamed Item ⋮ A Review of Data‐Driven Discovery for Dynamic Systems ⋮ Operator inference with roll outs for learning reduced models from scarce and low-quality data ⋮ A Variational Neural Network Approach for Glacier Modelling with Nonlinear Rheology ⋮ A deep neural network-based method for solving obstacle problems ⋮ Reconstruction of delay differential equations via learning parameterized dictionary ⋮ Deep neural network based adaptive learning for switched systems ⋮ Forecasting of nonlinear dynamics based on symbolic invariance ⋮ On the structure of time-delay embedding in linear models of non-linear dynamical systems ⋮ Sparse identification of truncation errors ⋮ Discovery of Dynamics Using Linear Multistep Methods ⋮ Weak SINDy: Galerkin-Based Data-Driven Model Selection ⋮ On the comparison of LES data-driven reduced order approaches for hydroacoustic analysis ⋮ Variational Inference Formulation for a Model-Free Simulation of a Dynamical System with Unknown Parameters by a Recurrent Neural Network ⋮ Construction of reduced-order models for fluid flows using deep feedforward neural networks ⋮ Data-Driven Learning of Nonautonomous Systems ⋮ Multiresolution convolutional autoencoders ⋮ Robust modeling of unknown dynamical systems via ensemble averaged learning ⋮ Spatiotemporally dynamic implicit large eddy simulation using machine learning classifiers

Uses Software

• L-BFGS-B
• L-BFGS
• LBFGS-B
• SINDy
• VAMPnets
• torchdiffeq

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• 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
• Linear system identification via an asymptotically stable observer
• 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
• Neural network modeling for near wall turbulent flow.
• The immersed boundary method: a projection approach
• A proposal on machine learning via dynamical systems
• On dynamic mode decomposition: theory and applications
• Nonlinear System Identification
• Dynamic mode decomposition of numerical and experimental data
• Numerical Gaussian Processes for Time-Dependent and Nonlinear Partial Differential Equations
• Discrete mechanics and variational integrators
• Recursive form of the eigensystem realization algorithm for system identification
• Non-linear system identification using neural networks
• Automated reverse engineering of nonlinear dynamical systems
• An eigensystem realization algorithm for modal parameter identification and model reduction
• Representations of non-linear systems: the NARMAX model
• Algorithm 778: L-BFGS-B
• A hierarchy of low-dimensional models for the transient and post-transient cylinder wake
• Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks
• Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
• Extracting Sparse High-Dimensional Dynamics from Limited Data
• Data-Driven Discovery of Closure Models
• Using machine learning to replicate chaotic attractors and calculate Lyapunov exponents from data
• Sparse reduced-order modelling: sensor-based dynamics to full-state estimation
• Model selection for dynamical systems via sparse regression and information criteria
• Learning partial differential equations via data discovery and sparse optimization
• Constrained sparse Galerkin regression

This page was built for publication: Deep learning of dynamics and signal-noise decomposition with time-stepping constraints