Analysing the Performance of Different Population Structures for an Agent-Based Evolutionary Algorithm

The Evolvable Agent model is a Peer-to-Peer Evolutionary Algorithm [4] which focuses on distributed optimisation over Peer-to-Peer infrastructures [7]. The main idea of the model is that every agent (i.e. individual) is designated as a peer (i.e. network node) and adopts a decentralised population structure defined by the underlying Peer-to-Peer protocol newscast [3]. That way, the population structure acquires a small network diameter which allows a fast dissemination of the best solutions. Additionally, speed of propagation holds with scaling network sizes due to the logarithmic growth of the network diameter.