Controlling mean exit time of stochastic dynamical systems based on quasipotential and machine learning

DOI10.1016/J.CNSNS.2023.107425MaRDI QIDQ6058680zbMATH OpenOpenAlexFDO

Authors Yang Li, Shenglan Yuan, Shengyuan Xu

Publication date 1 November 2023

Published in Communications in Nonlinear Science and Numerical Simulation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2209.13098

zbMATH Keywords

machine learningmean exit timequasipotentialstochastic control

Mathematics Subject Classification ID

Approximation methods and numerical treatment of dynamical systems (37M99) Numerical solutions to stochastic differential and integral equations (65C30) Learning and adaptive systems in artificial intelligence (68T05) Artificial neural networks and deep learning (68T07)

Abstract: The mean exit time escaping basin of attraction in the presence of white noise is of practical importance in various scientific fields. In this work, we propose a strategy to control mean exit time of general stochastic dynamical systems to achieve a desired value based on the quasipotential concept and machine learning. Specifically, we develop a neural network architecture to compute the global quasipotential function. Then we design a systematic iterated numerical algorithm to calculate the controller for a given mean exit time. Moreover, we identify the most probable path between metastable attractors with help of the effective Hamilton-Jacobi scheme and the trained neural network. Numerical experiments demonstrate that our control strategy is effective and sufficiently accurate.

Recommendations

• Control of exit time for Lagrangian systems with weak noise
• A machine learning method for computing quasi-potential of stochastic dynamical systems
• Hybrid Systems: Computation and Control
• Stochastic Control and Exit Probabilities of Jump Processes
• Stochastic exit time problems arising in process control
• Linear Exponential Quadratic Control for Mean Field Stochastic Systems
• Finite-time mean-nonovershooting control for stochastic nonlinear systems
• Control of dynamics via identical time-lagged stochastic inputs
• Mean exit time for stochastic dynamical systems driven by tempered stable Lévy fluctuations
• A quasi-sure approach to the control of non-Markovian stochastic differential equations

Cites work

• A data-driven approach for discovering stochastic dynamical systems with non-Gaussian Lévy noise
• A fast marching level set method for monotonically advancing fronts.
• A machine learning method for computing quasi-potential of stochastic dynamical systems
• A Primer on Noise-Induced Transitions in Applied Dynamical Systems
• An ordered line integral method for computing the quasi-potential in the case of variable anisotropic diffusion
• Analysis of a stochastic predator-prey model with prey subject to disease and Lévy noise
• Computing the quasipotential for nongradient SDEs in 3D
• Convergence rate for the ordered upwind method
• Extinction rates for fluctuation-induced metastabilities: a real-space WKB approach
• Finding the quasipotential for nongradient SDEs
• scientific article; zbMATH DE number 2006037 (Why is no real title available?)
• scientific article; zbMATH DE number 2027974 (Why is no real title available?)
• Hyperbolic periodic orbits in nongradient systems and small-noise-induced metastable transitions
• Invariant foliations for stochastic dynamical systems with multiplicative stable Lévy noise
• Limiting Exit Location Distributions in the Stochastic Exit Problem
• Modulation and amplitude equations on bounded domains for nonlinear SPDEs driven by cylindrical \(\alpha\)-stable Lévy processes
• Most probable dynamics of stochastic dynamical systems with exponentially light jump fluctuations
• Ordered Upwind Methods for Static Hamilton--Jacobi Equations: Theory and Algorithms
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Random Perturbations of Dynamical Systems
• Slow manifolds for a nonlocal fast-slow stochastic system with stable Lévy noise
• Slow manifolds for dynamical systems with non-Gaussian stable Lévy noise
• Stochastic bifurcation for two-time-scale dynamical system with -stable Lévy noise
• Stochastic bifurcation in single-species model induced by α-stable Lévy noise
• Stochastic bifurcations and tipping phenomena of insect outbreak systems driven by α-stable Lévy processes
• Stochastic turbulence for Burgers equation driven by cylindrical Lévy process
• The geometric minimum action method: A least action principle on the space of curves
• Using control to shape stochastic escape and switching dynamics

Cited in

(2)

• Rare events in a stochastic vegetation-water dynamical system based on machine learning
• Effective dynamics of interfaces for nonlinear SPDEs driven by multiplicative white noise

This page was built for publication: Controlling mean exit time of stochastic dynamical systems based on quasipotential and machine learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6058680)