Dimensional criterion for forecasting nonlinear systems by reservoir computing
This provides a practical criterion for selecting reservoir computing methods in forecasting nonlinear systems, though it is incremental as it refines existing optimization approaches.
The study tackled the problem of optimizing reservoir computer (RC) performance for forecasting chaotic systems by identifying a fractal dimension threshold (d ≈ 5.5) that determines whether unconnected or connected reservoir topologies are more effective, with results showing unconnected nodes outperform for systems with d ≲ 5.5 and connected nodes for d > 5.5.
Reservoir computers (RC) have proven useful as surrogate models in forecasting and replicating systems of chaotic dynamics. The quality of surrogate models based on RCs is crucially dependent on their optimal implementation that involves selecting optimal reservoir topology and hyperparameters. By systematically applying Bayesian hyperparameter optimization and using ensembles of reservoirs of various topology we show that connectednes of reservoirs is of significance only in forecasting and replication of chaotic system of sufficient complexity. By applying RCs of different topology in forecasting and replicating the Lorenz system, a coupled Wilson-Cowan system, and the Kuramoto-Sivashinsky system, we show that simple reservoirs of unconnected nodes (RUN) outperform reservoirs of connected nodes for target systems whose estimated fractal dimension dimension is $d \lesssim 5.5$ and that linked reservoirs are better for systems with $d > 5.5$. This finding is highly important for evaluation of reservoir computing methods and on selecting a method for prediction of signals measured on nonlinear systems.