UNIFICATION AND DIVERSITY OF COMPUTATION MODELS FOR GENERALIZED SWARM INTELLIGENCE

Swarm intelligence is a kind of nature-inspired heuristic optimization technique. Different computation models usually take on relative uniform characteristic though they usually have distinct extrinsic forms. These intelligent algorithms are coupling with deterministic and stochastic, the contradiction between necessity and accidental unity, which promotes the evolution of the inheritance and the creative process: stochastic is adopted to give creative ability to the implemented intelligent system, and a succession of certainty is acted to ensure the system is converging. In this paper, a computing framework of generalized swarm intelligence is proposed based on the unifying idea. The unified hiberarchy model and formalization description for swarm intelligence are represented. Several typical swarm intelligence algorithms, such as ant colony system (ACS), particle swarm optimization (PSO), estimation of distribution algorithms (EDA) and artificial immune algorithms (AIA) are addressed to validate the uniform idea of swarm intelligence, respectively.