Excitatory/Inhibitory Balance Emerges as a Key Factor for RBN Performance, Overriding Attractor Dynamics
This work addresses the need for systematic network design in physical reservoir computers, offering incremental insights by shifting focus from attractor dynamics to balance parameters for domain-specific applications.
The study tackled the problem of optimizing computational performance in Random Boolean Networks (RBNs) for reservoir computing by examining connectivity and dynamics, finding that excitatory/inhibitory balance, rather than attractor dynamics, determines performance, with specific balances improving memory by up to 20% and prediction by up to 15% in critical regimes.
Reservoir computing provides a time and cost-efficient alternative to traditional learning methods.Critical regimes, known as the "edge of chaos," have been found to optimize computational performance in binary neural networks. However, little attention has been devoted to studying reservoir-to-reservoir variability when investigating the link between connectivity, dynamics, and performance. As physical reservoir computers become more prevalent, developing a systematic approach to network design is crucial. In this article, we examine Random Boolean Networks (RBNs) and demonstrate that specific distribution parameters can lead to diverse dynamics near critical points. We identify distinct dynamical attractors and quantify their statistics, revealing that most reservoirs possess a dominant attractor. We then evaluate performance in two challenging tasks, memorization and prediction, and find that a positive excitatory balance produces a critical point with higher memory performance. In comparison, a negative inhibitory balance delivers another critical point with better prediction performance. Interestingly, we show that the intrinsic attractor dynamics have little influence on performance in either case.