A Framework for Evolving Spiking Neural P Systems

摘要

In the literature for spiking neural P systems (SN P systems) there is a need for further research on their optimization and, consequently, for automating this optimization process. We address this gap by designing a genetic algorithm (GA) framework that transforms an initial SN P system Pi(init, )designed to approximate a function f(w, x, y, ...) = z, into a smaller or more precise system Pi(final) that also approximates the output z given the same input/s w, x, y, ... The GA framework is con- strained to evolve only through its topology. The rules inside the neurons must stay constant, while the synapses and neurons may vary. Experiments conducted showed that using GA to evolve the topology of a designed Pi(init) decreases the number of its neurons and synapses, and makes it more precise. The GA framework is especially useful on SN P systems containing, as a subgraph of its synapse graph, a smaller SN P system computing f.