Peer-Reviewed Research

Constraining chaos: Enforcing dynamical invariants in the training of reservoir computers

Sofar Ocean

This paper was written by J. A. Platt, S. G. Penny, T. A. Smith, T. C. Chen, and H. D. I. Abarbanela.


Drawing on ergodic theory, a novel training method is introduced for machine learning-based forecasting methods for chaotic dynamical systems. The training enforces dynamical invariants—such as the Lyapunov exponent spectrum and fractal dimension—in the systems of interest, enabling longer and more stable forecasts when operating with limited data.

Sofar in the News
In the News
October 23, 2023
Wayfinder delivers 8% total equivalent fuel savings on Brazil-China (BDI C14) voyages arriving in early October
In the News
October 2, 2023
What was different about the latest Advanced Naval Technology Exercise
In the News
September 25, 2023
Sofar Ocean's Superior Weather Forecasts Minimize Voyage Costs

Related Stories