Inferring the dynamics of oscillatory systems using recurrent neural networks

DOI10.1063/1.5096918MaRDI QIDQ5227599zbMATH OpenOpenAlexWikidataFDO

Authors Rok Cestnik, Markus Abel

Publication date 6 August 2019

Published in Chaos: An Interdisciplinary Journal of Nonlinear Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1904.03026

Mathematics Subject Classification ID

Strange attractors, chaotic dynamics of systems with hyperbolic behavior (37D45) Nonlinear oscillations and coupled oscillators for ordinary differential equations (34C15)

Abstract: We investigate the predictive power of recurrent neural networks for oscillatory systems not only on the attractor, but in its vicinity as well. For this we consider systems perturbed by an external force. This allows us to not merely predict the time evolution of the system, but also study its dynamical properties, such as bifurcations, dynamical response curves, characteristic exponents etc. It is shown that they can be effectively estimated even in some regions of the state space where no input data were given. We consider several different oscillatory examples, including self-sustained, excitatory, time-delay and chaotic systems. Furthermore, with a statistical analysis we assess the amount of training data required for effective inference for two common recurrent neural network cells, the long short-term memory and the gated recurrent unit.

Recommendations

• A recurrent neural network for modelling dynamical systems
• Approximation of dynamical systems by recurrent neural networks
• scientific article; zbMATH DE number 1190938
• scientific article; zbMATH DE number 1304847
• scientific article; zbMATH DE number 5139255
• Modeling of continuous time dynamical systems with input by recurrent neural networks
• scientific article; zbMATH DE number 2221002
• Dynamical Systems–Based Neural Networks

Cites work

• scientific article; zbMATH DE number 3728070 (Why is no real title available?)
• A practical method for calculating largest Lyapunov exponents from small data sets
• A recurrent neural network for modelling dynamical systems
• An equation for continuous chaos
• Deep learning
• Determining Lyapunov exponents from a time series
• Differentiable manifolds
• Floquet theory for partial differential equations
• NONPARAMETRIC MODELING AND SPATIOTEMPORAL DYNAMICAL SYSTEMS
• Nonlinear climate dynamics
• Nonlinear oscillations, dynamical systems, and bifurcations of vector fields
• On the numerical computation of Poincaré maps
• Rhythms of the brain.
• Synchronization: a universal concept in nonlinear sciences
• Using machine learning to replicate chaotic attractors and calculate Lyapunov exponents from data

Cited in

(8)

• Deep learning-based state prediction of the Lorenz system with control parameters
• Oscillatory neural network for pattern recognition: trajectory based classification and supervised learning
• Revealing hidden dynamics from time-series data by ODENet
• A recurrent neural network for modelling dynamical systems
• scientific article; zbMATH DE number 5139255 (Why is no real title available?)
• Learning brain dynamics with coupled low-dimensional nonlinear oscillators and deep recurrent networks
• Phase-amplitude coordinate-based neural networks for inferring oscillatory dynamics
• ODE-RU: a dynamical system view on recurrent neural networks

Describes a project that uses

Uses Software

• TensorFlow

This page was built for publication: Inferring the dynamics of oscillatory systems using recurrent neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5227599)