Learning Spatiotemporal Chaos Using Next-Generation Reservoir Computing

TL;DR

This paper presents a novel reservoir computing approach for predicting high-dimensional spatiotemporal chaos, achieving state-of-the-art accuracy with significantly reduced training time and data requirements by leveraging translational symmetry.

Contribution

The authors introduce a next-generation reservoir computing architecture that outperforms existing methods in speed and data efficiency for spatiotemporal chaos prediction.

Findings

Training time reduced by 10^3-10^4 times

Training data set size reduced by ~100 times

Computational cost decreased by a factor of ~10

Abstract

Forecasting the behavior of high-dimensional dynamical systems using machine learning requires efficient methods to learn the underlying physical model. We demonstrate spatiotemporal chaos prediction using a machine learning architecture that, when combined with a next-generation reservoir computer, displays state-of-the-art performance with a computational time $10^3-10^4$ times faster for training process and training data set $\sim 10^2$ times smaller than other machine learning algorithms. We also take advantage of the translational symmetry of the model to further reduce the computational cost and training data, each by a factor of $\sim$10.