Learning Stochastic Dynamical System via Flow Map Operator

DOI10.1016/J.JCP.2024.112984OpenAlexW4394567136MaRDI QIDQ6435560FDOQ6435560

Authors Y. Chen, Dongbin Xiu 

Publication date 5 May 2023

Full work available at URL https://doi.org/10.1016/j.jcp.2024.112984

Mathematics Subject Classification ID

Artificial intelligence (68Txx) Numerical methods for ordinary differential equations (65Lxx) Numerical methods for partial differential equations, boundary value problems (65Nxx)

Abstract: We present a numerical framework for learning unknown stochastic dynamical systems using measurement data. Termed stochastic flow map learning (sFML), the new framework is an extension of flow map learning (FML) that was developed for learning deterministic dynamical systems. For learning stochastic systems, we define a stochastic flow map that is a superposition of two sub-flow maps: a deterministic sub-map and a stochastic sub-map. The stochastic training data are used to construct the deterministic sub-map first, followed by the stochastic sub-map. The deterministic sub-map takes the form of residual network (ResNet), similar to the work of FML for deterministic systems. For the stochastic sub-map, we employ a generative model, particularly generative adversarial networks (GANs) in this paper. The final constructed stochastic flow map then defines a stochastic evolution model that is a weak approximation, in term of distribution, of the unknown stochastic system. A comprehensive set of numerical examples are presented to demonstrate the flexibility and effectiveness of the proposed sFML method for various types of stochastic systems.

Recommendations

• Stochastic physics-informed neural ordinary differential equations
• DynGMA: a robust approach for learning stochastic differential equations from data
• SDYN-GANs: adversarial learning methods for multistep generative models for general order stochastic dynamics
• One-shot learning of stochastic differential equations with data adapted kernels

Cited In (7)

• DynGMA: a robust approach for learning stochastic differential equations from data
• Stochastic dynamics and data science
• Deep-OSG: deep learning of operators in semigroup
• Meta-state-space learning: an identification approach for stochastic dynamical systems
• Transition path properties for one-dimensional non-Markovian models
• Reservoir computing with error correction: long-term behaviors of stochastic dynamical systems
• SDYN-GANs: adversarial learning methods for multistep generative models for general order stochastic dynamics

This page was built for publication: Learning Stochastic Dynamical System via Flow Map Operator

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6435560)