A 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 nonidentical phase oscillators, a heterogeneous complex network of nonidentical phase oscillators, and a heterogeneous complex network of nonidentical chaotic oscillators.

Close

PrevNext

View figure in article

• Machine learning
• Neuromorphic computing
• Chaotic systems
• Complex networks
• Coupled oscillators
• Machine learning models
• Time series analysis