Versatile Reservoir Computing for Heterogeneous Complex Networks

TL;DR

This paper introduces a versatile reservoir computing approach capable of modeling and maintaining the dynamics of heterogeneous complex networks, even with limited training data, across various network types.

Contribution

It presents a novel reservoir computing scheme that can replicate and preserve the dynamics of large heterogeneous networks using minimal training.

Findings

Single reservoir trained on subset can replicate any element's dynamics.

The scheme preserves network dynamics over finite time horizons.

Validated on three diverse complex network models.

Abstract

A new machine learning scheme, termed versatile reservoir computing, is proposed for sustaining the dynamics of heterogeneous complex networks. We show that a single, small-scale reservoir computer trained on time series from a subset of elements is able to replicate the dynamics of any element in a large-scale complex network, though the elements are of different intrinsic parameters and connectivities. Furthermore, by substituting failed elements with the trained machine, we demonstrate that the collective dynamics of the network can be preserved accurately over a finite time horizon. The capability and effectiveness of the proposed scheme are validated on three representative network models: a homogeneous complex network of non-identical phase oscillators, a heterogeneous complex network of non-identical phase oscillators, and a heterogeneous complex network of non-identical chaotic oscillators.