Reservoir Computing Using Nonuniform Binary Cellular Automata

Abstract

The reservoir computing (RC) paradigm utilizes a dynamical system (a reservoir) and a linear classifier (a readout layer) to process data from sequential classification tasks. In this paper, the usage of cellular automata (CAs) as a reservoir is investigated. The use of CAs in RC has been showing promising results. In this paper, it is shown that some cellular automaton (CA) rules perform better than others and the reservoir performance is improved by increasing the size of the CA reservoir itself. In addition, the usage of parallel loosely coupled (nonuniform) CA reservoirs, where each reservoir has a different CA rule, is investigated. The experiments performed on nonuniform CA reservoirs provide valuable insights into CA reservoir design. The results herein show that some rules do not work well together, while other combinations work remarkably well. This suggests that nonuniform CAs could represent a powerful tool for novel CA reservoir implementations.