The effect of noise on the performance of cultural evolution in multi-agent systems

This work examines the effect of the addition of noise to the cultural learning process of a population of agents. Experiments are undertaken using an artificial life simulator capable of simulating population learning (through the use of genetic algorithms) and lifetime learning (through the use of neural networks). To simulate cultural learning, (the exchange of information through nongenetic means) a group of highly fit agents is selected at each generation to function as teachers which are assigned a number of pupils to instruct. Cultural exchanges occur through a hidden layer of an agent´s neural network known as the verbal layer. Through the use of back-propagation, a pupil agent imitates the teacher´s behaviour and overall population fitness is increased. We introduce cultural mutation into a population of agents by adding noise to cultural exchanges between teacher and pupil agents. We conduct a series of experiments with varying values of cultural mutation to study the effects of this operator on the performance of the population. We show that the addition of noise to cultural exchanges can improve on the performance of cultural learning.