A framework for machine learning of model error in dynamical systems
DOI10.1090/cams/10arXiv2107.06658MaRDI QIDQ6076655
Matthew E. Levine, Andrew M. Stuart
Publication date: 17 October 2023
Published in: Communications of the American Mathematical Society (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/2107.06658
dynamical systems; recurrent neural networks; statistical learning; model error; reservoir computing; random features
68T30: Knowledge representation
65L05: Numerical methods for initial value problems involving ordinary differential equations
37A30: Ergodic theorems, spectral theory, Markov operators
37M10: Time series analysis of dynamical systems
41A30: Approximation by other special function classes
37M25: Computational methods for ergodic theory (approximation of invariant measures, computation of Lyapunov exponents, entropy, etc.)
Related Items
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Bagging predictors
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Analysis of the 3DVAR filter for the partially observed Lorenz '63 model
- Reservoir computing approaches to recurrent neural network training
- Bayesian solution uncertainty quantification for differential equations
- Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism
- Modeling of missing dynamical systems: deriving parametric models using a nonparametric framework
- Modelling and control of dynamic systems using Gaussian process models
- A family of embedded Runge-Kutta formulae
- Neural networks and dynamical systems
- Statistical inference for ergodic diffusion processes.
- Data-based stochastic model reduction for the Kuramoto-Sivashinsky equation
- A dynamic subgrid scale model for large eddy simulations based on the Mori-Zwanzig formalism
- Statistical and computational inverse problems.
- A computational strategy for multiscale systems with applications to Lorenz 96 model
- Machine-learning error models for approximate solutions to parameterized systems of nonlinear equations
- Model reduction and neural networks for parametric PDEs
- Error bounds of the invariant statistics in machine learning of ergodic Itô diffusions
- Learning dynamical systems from data: a simple cross-validation perspective. I: Parametric kernel flows
- Supervised learning from noisy observations: combining machine-learning techniques with data assimilation
- Kernel-based prediction of non-Markovian time series
- Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network
- Data-driven model reduction, Wiener projections, and the Koopman-Mori-Zwanzig formalism
- Operator-theoretic framework for forecasting nonlinear time series with kernel analog techniques
- Machine learning for prediction with missing dynamics
- Solving and learning nonlinear PDEs with Gaussian processes
- Stable \textit{a posteriori} LES of 2D turbulence using convolutional neural networks: backscattering analysis and generalization to higher \(Re\) via transfer learning
- Modelling spatiotemporal dynamics from Earth observation data with neural differential equations
- Echo state networks are universal
- Variance continuity for Lorenz flows
- A priori estimates of the population risk for two-layer neural networks
- Data-driven spectral analysis of the Koopman operator
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Consistency of maximum likelihood estimation for some dynamical systems
- On dynamic mode decomposition: theory and applications
- Numerical techniques for multi-scale dynamical systems with stochastic effects
- The Random Feature Model for Input-Output Maps between Banach Spaces
- Optimal prediction and the Mori–Zwanzig representation of irreversible processes
- On the estimation of the Mori-Zwanzig memory integral
- A priori estimation of memory effects in reduced-order models of nonlinear systems using the Mori–Zwanzig formalism
- Extracting Sparse High-Dimensional Dynamics from Limited Data
- Using machine learning to replicate chaotic attractors and calculate Lyapunov exponents from data
- Stable architectures for deep neural networks
- Deterministic Nonperiodic Flow
- Orthogonal Gaussian process models
- Kernel Analog Forecasting: Multiscale Test Problems
- Augmenting physical models with deep networks for complex dynamics forecasting*
- Learning stochastic closures using ensemble Kalman inversion
- Autodifferentiable Ensemble Kalman Filters
- The Discovery of Dynamics via Linear Multistep Methods and Deep Learning: Error Estimation
- Extracting Structured Dynamical Systems Using Sparse Optimization With Very Few Samples
- Learning latent dynamics for partially observed chaotic systems
- Discovery of Dynamics Using Linear Multistep Methods
- SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics
- Model Reduction with Memory and the Machine Learning of Dynamical Systems
- Probabilistic Forecasting and Bayesian Data Assimilation
- Data-driven discovery of coordinates and governing equations
- Data Assimilation
- Data Assimilation
- Learning representations by back-propagating errors
- Turbulence Modeling in the Age of Data
- Central limit theorems and invariance principles for Lorenz attractors
- Multiscale Methods
- On Information and Sufficiency
- Inverse acoustic and electromagnetic scattering theory
- Smoothing noisy data with spline functions
- Approximation by superpositions of a sigmoidal function