Learning spatiotemporal chaos using next-generation reservoir computing
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 103–104 times faster for training process and training data set ∼102 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 ∼10.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Sep 01, 2022
- Source ID
- 10.1063/5.0098707
Entities
People
- Daniel J Gauthier
- Wendson A. S. Barbosa
Organizations
- Air Force Office of Scientific Research
- Ohio State University