Predicting complex system changes by combining learned models
Extrapolating learned dynamics across bifurcations based on generalized synchronization
Machine Learning
Summary
Predicting how complex systems behave when they suddenly change is hard. The authors found a way to use multiple trained neural network parts together to guess what happens in new situations they haven't seen before. Their math shows this guessing can be very accurate. They tested their approach and proved it works for chaotic systems that change behavior sharply.
What this means in practice
- •For physics simulation teams: Predict the future behavior of physical systems undergoing sudden qualitative changes by combining existing trained dynamics models.
- •For climate modelers: Improve forecasting models for climate systems by extrapolating learned patterns across critical transition points in dynamical behavior.
Authors
Lingfeng Chen, Masanobu Inubushi
Abstract
Predicting dynamics across bifurcations by extrapolating trained models is crucial in diverse fields of physics. Generalized synchronization provides a framework for understanding dynamical reconstruction by neural networks, especially in the context of reservoir computing. Building on this framework, we derive a formula combining pretrained readouts to reconstruct unseen chaotic dynamics. Using classical Lagrange interpolation theory, we derive rigorous bounds on readout errors and conjugacy residuals. Numerical examples demonstrate accurate prediction of chaotic dynamics across bifurcations.