Estimation of Lyapunov Exponents with use of Automatic Differentiation

Published: 27 February 2025| Version 1 | DOI: 10.17632/9k4n5m9d6f.1
Contributors:
,

Description

The dataset contains Python scripts connected with estimation of Lyapunov Exponents with use of Automatic Differentiation. They show that: a) Automatic Differentiation (AD) can be successfully applied in Lyapunov Exponents (LEs) estmination; b) AD approach is slighlty less accurate then the one based on the variational equation; b) AD approach is less efficient then the one based on the variational equation.

Files

Steps to reproduce

Scripts have been tested using Python 3.12.5 and JAX 0.4.37, which can be installed using "pip" package manager.

Institutions

Politechnika Lodzka Wydzial Mechaniczny

Categories

Nonlinear Dynamical Systems, Chaos Theory, Computational Method in Nonlinear Dynamics

Funding

Lodz University of Technology

RNCWN/WFS/6/2024

Licence