Reservoir computing with simple oscillators: Virtual and real networks

Abstract

The reservoir computing scheme is a machine learning mechanism which utilizes the naturally occuring computational capabilities of dynamical systems. One important subset of systems that has proven powerful both in experiments and theory are delay-systems. In this work, we investigate the reservoir computing performance of hybrid network-delay systems systematically by evaluating the NARMA10 and the Sante Fe task.. We construct 'multiplexed networks' that can be seen as intermediate steps on the scale from classical networks to the 'virtual networks' of delay systems. We find that the delay approach can be extended to the network case without loss of computational power, enabling the construction of faster reservoir computing substrates.

References in corpus (5)

• High performance photonic reservoir computer based on a coherently driven passive cavity
• Amplitude-phase coupling drives chimera states in globally coupled laser networks
• Online Training of an Opto-Electronic Reservoir Computer Applied to Real-Time Channel Equalisation
• Modulating coherence resonance in non-excitable systems by time-delayed feedback
• Bistability in Two Simple Symmetrically Coupled Oscillators with Symmetry-broken Amplitude- and Phase-Locking

Cited by in corpus (7)

• Irreversibility, heat and information flows induced by non-reciprocal interactions
• Performance boost of time-delay reservoir computing by non-resonant clock cycle
• Insight into Delay Based Reservoir Computing via Eigenvalue Analysis
• Limitations of the recall capabilities in delay based reservoir computing systems
• Step-like dependence of memory function on pulse width in spintronics reservoir computing
• Dynamical Systems as Temporal Feature Spaces
• Non-reciprocal hidden degrees of freedom: A unifying perspective on memory, feedback, and activity