Steps Towards a Quantitative Analysis of Individuality and its Maintenance: A Case Study with Multi-Agent Systems

摘要

In the context of the study of the evolution of new levels of biological organisation, we present initial work aimed at constructing a metric to detect individuality, or cooperative systems of agents. The metric is based upon the idea that individuals must possess a distinct, reduced set of higher-level variables with respect to the degrees of freedom of their components. Further, in biology, each new level of organisation is a homeostatic entity in its own right, and we argue that the origin of homeostasis amongst a group of entities is the critical point at which they become a new, higher-level, system. We investigate a new measure of the degree of individuality in collective systems of agents, based on calculating the effective number of degrees of freedom in a system, and how this changes with increasing external perturbation. The results of applying the measure to collectives of simulated Kheperas are presented, and the applications to homeostatic and non-homeostatic systems are discussed.