Controlling dynamical systems to complex target states using machine learning: next-generation vs. classical reservoir computing

TL;DR

This paper compares classical and next-generation reservoir computing for controlling complex dynamical systems, showing that next-generation RC performs better with limited training data, expanding practical control applications.

Contribution

It demonstrates that next-generation reservoir computing outperforms classical methods in data-scarce scenarios for controlling nonlinear dynamical systems.

Findings

Classical reservoir computing excels with ample training data.

Next-generation reservoir computing outperforms in limited data conditions.

Both methods perform similarly with sufficient training data.

Abstract

Controlling nonlinear dynamical systems using machine learning allows to not only drive systems into simple behavior like periodicity but also to more complex arbitrary dynamics. For this, it is crucial that a machine learning system can be trained to reproduce the target dynamics sufficiently well. On the example of forcing a chaotic parametrization of the Lorenz system into intermittent dynamics, we show first that classical reservoir computing excels at this task. In a next step, we compare those results based on different amounts of training data to an alternative setup, where next-generation reservoir computing is used instead. It turns out that while delivering comparable performance for usual amounts of training data, next-generation RC significantly outperforms in situations where only very limited data is available. This opens even further practical control applications in real world problems where data is restricted.