Dynamics-Informed Deep Learning for Predicting Extreme Events

TL;DR

This paper introduces a data-driven, mechanism-aware forecasting framework that uses reduced-order models and Transformer-based prediction to improve long-term prediction of rare extreme events in chaotic systems.

Contribution

It develops an interpretable, efficient method to identify transient instabilities as precursors for extreme events without needing the governing equations.

Findings

Extended prediction horizons for extreme events.

Efficient instability detection using low-dimensional subspaces.

Improved forecasting accuracy over baseline methods.

Abstract

Predicting extreme events in high-dimensional chaotic dynamical systems remains a fundamental challenge, as such events are rare, intermittent, and arise from transient dynamical mechanisms that are difficult to infer from limited observations. Accordingly, real-time forecasting calls for precursors that encode the mechanisms driving extremes, rather than relying solely on statistical associations. We propose a fully data-driven framework for long-lead prediction of extreme events that constructs interpretable, mechanism-aware precursors by explicitly tracking transient instabilities preceding event onset. The approach leverages a reduced-order formulation to compute finite-time Lyapunov exponent (FTLE)-like precursors directly from state snapshots, without requiring knowledge of the governing equations. To avoid the prohibitive computational cost of classical FTLE computation, instability growth is evaluated in an adaptively evolving low-dimensional subspace spanned by Optimal Time-Dependent (OTD) modes, enabling efficient identification of transiently amplifying directions. These precursors are then provided as input to a Transformer-based model, enabling forecast of extreme event observables. We demonstrate the framework on Kolmogorov flow, a canonical model of intermittent turbulence. The results show that explicitly encoding transient instability mechanisms substantially extends practical prediction horizons compared to baseline observable-based approaches.