One-shot learning of stochastic differential equations with data adapted kernels

DOI10.1016/j.physd.2022.133583OpenAlexW4308922332MaRDI QIDQ2111726

Houman Owhadi, Matthieu Darcy, Boumediene Hamzi, Giulia Livieri, Peyman Tavallali

Publication date: 17 January 2023

Published in: Physica D (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2209.12086

zbMATH Keywords

Gaussian processesstochastic differential equationskernel methodsmachine learningcomputational graph completiontimes series forecasting

Mathematics Subject Classification ID

Computer science (68-XX) Systems theory; control (93-XX)

Related Items

Learning dynamical systems from data: a simple cross-validation perspective. IV: Case with partial observations, Gaussian process hydrodynamics

Uses Software

• Scikit
• Spearmint
• Hyperopt
• SMAC

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Discovering governing equations from data by sparse identification of nonlinear dynamical systems
• SMAC
• Maximum likelihood estimation of drift and diffusion functions
• Nonlinear prediction of chaotic time series
• The Fokker-Planck equation. Methods of solution and applications.
• Kernel methods for the approximation of some key quantities of nonlinear systems
• Analysis of data sets of stochastic systems
• Dimensionality reduction of complex metastable systems via kernel embeddings of transition manifolds
• Kernel methods for center manifold approximation and a weak data-based version of the center manifold theorem
• Learning dynamical systems from data: a simple cross-validation perspective. I: Parametric kernel flows
• Data-driven approximation of the Koopman generator: model reduction, system identification, and control
• Operator-theoretic framework for forecasting nonlinear time series with kernel analog techniques
• Solving and learning nonlinear PDEs with Gaussian processes
• Kernel flows: from learning kernels from data into the abyss
• Computational graph completion
• On the mathematical foundations of learning
• 9 Kernel methods for surrogate modeling
• THE RECONSTRUCTION OF THE FOKKER-PLANCK AND MASTER EQUATIONS ON THE BASIS OF EXPERIMENTAL DATA: H-THEOREM AND S-THEOREM
• Using machine learning to replicate chaotic attractors and calculate Lyapunov exponents from data
• A note on kernel methods for multiscale systems with critical transitions
• Kernel Methods for the Approximation of Nonlinear Systems
• Variational Inference for Stochastic Differential Equations