Towards unconventional computing through simulated evolution: Learning classifier system control of non-linear media
Larry Bull, Adam Budd, Christopher Stone, Ivan S. Uroukov, Ben de Lacy Costello, Andrew Adamatzky · 2007
Abstract. We propose that the behaviour of non-linear media can be controlled automatically through evolutionary learning. By extension, forms of unconventional computing, i.e., massively parallel non-linear computers, can be realised by such an approach. In this initial study a light-sensitive sub-excitable Belousov-Zhabotinski reaction in which a checkerboard image comprising of varying light intensity cells projected onto the surface of a thin silica gel impregnated with a catalyst and indicator is controlled using a Learning Classifier System, as is a form of in vitro neuronal network. In the former, pulses of wave fragments are injected into the checkerboard grid resulting in rich spatio-temporal behaviour and a Learning Classifier System is shown able to direct the fragments to an arbitrary position through dynamic control of the light intensity within each cell in both simulated and real chemical systems. Similarly, a Learning Classifier System is shown able to control the electrical stimulation of cultured neuronal networks such that they display elementary learning, i.e., so that they respond to a given input signal in a pre-specified way. Results indicate that the learned stimulation protocols identify seemingly fundamental properties of in vitro neuronal networks. Use of another learning scheme presented in the literature confirms these properties.