The Emergence of Social Complexity in Optimizing Mechanical Design Problems via Cultural Learning

摘要

Previous work on global constrained optimization problems demonstrated that knowledge swarms or meta-level interaction of knowledge sources in the belief space produced emergent cultural transmission. The swarming of individuals in the population space, Knowledge Swarms, has emerged as a result of these meta-level swarms in the belief space. The interaction of the knowledge sources in the belief space produced emergent phases of problem solving that reflected a branch and bound algorithmic process. These phases result in the emergence of individual roles within the population that leads to organized swarming in the population level and knowledge swarms in the social belief space. In this paper we describe a new framework based on Cultural Algorithms enhanced with heterogeneous social networks guided by belief knowledge in order to solve constrained mechanical design optimization problems. We use it to solve Golinski's Speed reducer problem here. We conclude by describing the emergent features that give cultural systems their power to learn and adapt.