Deep learning delay coordinate dynamics for chaotic attractors from partial observable data

TL;DR

This paper demonstrates how deep neural networks can learn delay coordinate maps to predict chaotic attractor dynamics from partial observations, providing an effective alternative to recurrent models.

Contribution

The authors introduce a deep learning approach to learn delay coordinate dynamics directly from partial data, enabling accurate chaos prediction without high-dimensional internal states.

Findings

Deep ANNs successfully predict Lorenz system chaos from scalar data.

The method extends to multivariate data, scaling with system complexity.

Predictions outperform traditional RNN models in efficiency and accuracy.

Abstract

A common problem in time series analysis is to predict dynamics with only scalar or partial observations of the underlying dynamical system. For data on a smooth compact manifold, Takens theorem proves a time delayed embedding of the partial state is diffeomorphic to the attractor, although for chaotic and highly nonlinear systems learning these delay coordinate mappings is challenging. We utilize deep artificial neural networks (ANNs) to learn discrete discrete time maps and continuous time flows of the partial state. Given training data for the full state, we also learn a reconstruction map. Thus, predictions of a time series can be made from the current state and several previous observations with embedding parameters determined from time series analysis. The state space for time evolution is of comparable dimension to reduced order manifold models. These are advantages over recurrent neural network models, which require a high dimensional internal state or additional memory terms and hyperparameters. We demonstrate the capacity of deep ANNs to predict chaotic behavior from a scalar observation on a manifold of dimension three via the Lorenz system. We also consider multivariate observations on the Kuramoto-Sivashinsky equation, where the observation dimension required for accurately reproducing dynamics increases with the manifold dimension via the spatial extent of the system.