Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networks

TL;DR

This paper presents a novel data-driven forecasting approach using LSTM neural networks for high-dimensional chaotic systems, demonstrating superior short-term prediction accuracy and a hybrid model for improved convergence.

Contribution

Introduces a hybrid LSTM-based forecasting method with a stochastic extension for high-dimensional chaotic systems, outperforming Gaussian processes in accuracy.

Findings

LSTM networks outperform Gaussian processes in short-term forecasting.

The hybrid MSM-LSTM model ensures convergence to the invariant measure.

The method is effective across different complex systems like Lorenz 96 and climate models.

Abstract

We introduce a data-driven forecasting method for high-dimensional chaotic systems using long short-term memory (LSTM) recurrent neural networks. The proposed LSTM neural networks perform inference of high-dimensional dynamical systems in their reduced order space and are shown to be an effective set of nonlinear approximators of their attractor. We demonstrate the forecasting performance of the LSTM and compare it with Gaussian processes (GPs) in time series obtained from the Lorenz 96 system, the Kuramoto-Sivashinsky equation and a prototype climate model. The LSTM networks outperform the GPs in short-term forecasting accuracy in all applications considered. A hybrid architecture, extending the LSTM with a mean stochastic model (MSM-LSTM), is proposed to ensure convergence to the invariant measure. This novel hybrid method is fully data-driven and extends the forecasting capabilities of LSTM networks.