Constraining chaos: Enforcing dynamical invariants in the training of reservoir computers
- 1 October 2023
- journal article
- research article
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (10)
Abstract
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. 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.Keywords
All Related Versions
Funding Information
- Office of Naval Research (N00014-19-1-2522)
- Office of Naval Research (N00014-20-1-2580)
- National Oceanic and Atmospheric Administration (NA20OAR4600277)
- Cooperative Institute for Research in Environmental Sciences (NA17OAR4320101)
This publication has 39 references indexed in Scilit:
- Comparison of Local Ensemble Transform Kalman Filter, 3DVAR, and 4DVAR in a Quasigeostrophic ModelMonthly Weather Review, 2009
- Chaotic mixing in microdropletsLab on a Chip, 2006
- Introduction: mixing in microfluidicsPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2004
- Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on PerturbationsNeural Computation, 2002
- The general problem of the stability of motionInternational Journal of Control, 1992
- Experimental evidence of chaotic itinerancy and spatiotemporal chaos in opticsPhysical Review Letters, 1990
- Ergodic theory of chaos and strange attractorsReviews of Modern Physics, 1985
- The liapunov dimension of strange attractorsJournal of Differential Equations, 1983
- Kolmogorov entropy and numerical experimentsPhysical Review A, 1976
- A New Approach to Linear Filtering and Prediction ProblemsJournal of Basic Engineering, 1960