Constraining chaos: Enforcing dynamical invariants in the training of reservoir computers
Journal Article
·
· Chaos: An Interdisciplinary Journal of Nonlinear Science
- Univ. of California, San Diego, CA (United States)
- Sofar Ocean, San Francisco, CA (United States); Univ. of Colorado, Boulder, CO (United States)
- Univ. of Colorado, Boulder, CO (United States); National Oceanic and Atmospheric Administration (NOAA), Boulder, CO (United States)
- Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
- Univ. of California, San Diego, CA (United States). Scripps Inst. of Oceanography
Drawing on ergodic theory, we introduce a novel training method for machine learning based forecasting methods for chaotic dynamical systems. The training enforces dynamical invariants—such as the Lyapunov exponent spectrum and the fractal dimension—in the systems of interest, enabling longer and more stable forecasts when operating with limited data. The technique is demonstrated in detail using reservoir computing, a specific kind of recurrent neural network. Finally, results are given for the Lorenz 1996 chaotic dynamical system and a spectral quasi-geostrophic model of the atmosphere, both typical test cases for numerical weather prediction.
- Research Organization:
- Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
- Sponsoring Organization:
- USDOE
- Grant/Contract Number:
- AC05-76RL01830
- OSTI ID:
- 2246611
- Report Number(s):
- PNNL-SA-191042
- Journal Information:
- Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 33, Issue 10; ISSN 1054-1500
- Publisher:
- American Institute of Physics (AIP)Copyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Resonances in a Chaotic Attractor Crisis of the Lorenz Flow
Neural Network Based Montioring and Control of Fluidized Bed.
A Non‐Intrusive Machine Learning Framework for Debiasing Long‐Time Coarse Resolution Climate Simulations and Quantifying Rare Events Statistics
Journal Article
·
Thu Feb 15 00:00:00 EST 2018
· Journal of Statistical Physics
·
OSTI ID:2246611
Neural Network Based Montioring and Control of Fluidized Bed.
Technical Report
·
Mon Apr 01 00:00:00 EST 1996
·
OSTI ID:2246611
A Non‐Intrusive Machine Learning Framework for Debiasing Long‐Time Coarse Resolution Climate Simulations and Quantifying Rare Events Statistics
Journal Article
·
Tue Mar 26 00:00:00 EDT 2024
· Journal of Advances in Modeling Earth Systems
·
OSTI ID:2246611
+3 more