Reservoir computing with simple oscillators: Virtual and real networks
This work addresses the need for more efficient reservoir computing substrates, offering incremental improvements for machine learning applications.
The study tackled the problem of enhancing reservoir computing performance by systematically evaluating hybrid network-delay systems on tasks like NARMA10 and Santa Fe, finding that extending delay approaches to networks maintains computational power and enables faster substrates.
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.