Data-Driven Identification of Attractors Using Machine Learning

doi:10.1137/25M1765547

Data-Driven Identification of Attractors Using Machine Learning

2026-01-01

0

PRE

AI

Mischaikow, Konstantin

T

Tatasciore, Paul

DOI：10.1137/25M1765547

Original

Original request for help

Save

Abstract

En 中文

In this paper we explore challenges in developing a topological framework in which machine learning can be used to robustly characterize global dynamics. Specifically, we focus on learning a useful discretization of the phase space of a flow on a compact, hyperrectangle in Rn from a neural network trained on labeled orbit data. A characterization of the structure of the global dynamics is obtained from approximations of attracting neighborhoods provided by the phase space discretization. The perspective that motivates this work is based on Conley's topological approach to dynamics, which provides a means to evaluate the efficacy and efficiency of our approach.

Keywords:

global dynamics

attracting neighborhood

machine learning

data-driven dynamics

Conley index

Data-Driven Identification of Attractors Using Machine Learning

Abstract

Journal

Organization