Evolution of spiking neural networks robust to noise and damage for control of simple animats

Abstract

One of the central questions of biology is how complex biological systems can continue functioning in the presence of perturbations, damage, and mutational insults. This paper investigates evolution of spiking neural networks, consisting of adaptive exponential neurons. The networks are encoded in linear genomes in a manner inspired by genetic networks. The networks control a simple animat, with two sensors and two actuators, searching for targets in a simple environment. The results show that the presence of noise on the membrane voltage during evolution allows for evolution of efficient control and robustness to perturbations to the value of the neural parameters of neurons.