Learning Stochastic Dynamics with Statistics-Informed Neural Network

DOI10.1016/J.JCP.2022.111819MaRDI QIDQ6392081arXivFDO

Authors Yuanran Zhu, Yu-Hang Tang, Chang Ho Kim

Publication date 24 February 2022

Published in arXiv

Full work available at URL https://doi.org/10.1016/J.JCP.2022.111819

Mathematics Subject Classification ID

Stochastic analysis (60Hxx) Artificial intelligence (68Txx) Time-dependent statistical mechanics (dynamic and nonequilibrium) (82Cxx)

arXiv classification

cond-mat.stat-mechcs.AIcs.LGmath.DSphysics.comp-ph

Abstract: We introduce a machine-learning framework named statistics-informed neural network (SINN) for learning stochastic dynamics from data. This new architecture was theoretically inspired by a universal approximation theorem for stochastic systems, which we introduce in this paper, and the projection-operator formalism for stochastic modeling. We devise mechanisms for training the neural network model to reproduce the correct emph{statistical} behavior of a target stochastic process. Numerical simulation results demonstrate that a well-trained SINN can reliably approximate both Markovian and non-Markovian stochastic dynamics. We demonstrate the applicability of SINN to coarse-graining problems and the modeling of transition dynamics. Furthermore, we show that the obtained reduced-order model can be trained on temporally coarse-grained data and hence is well suited for rare-event simulations.

This page was built for publication: Learning Stochastic Dynamics with Statistics-Informed Neural Network

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6392081)