Hierarchical Swarm Model: A New Approach to Optimization

摘要: This paper presents a novel optimization model called hierarchical swarm (HSO), which simulates the natural complex system from where more intelligence can emerge for problems solving. proposed is intended to suggest ways that performance of HSO-based algorithms on be significantly improved. improvement obtained by constructing HSO hierarchies, means an agent in higher level composed swarms other agents lower and different levels evolve spatiotemporal scale. A algorithm (named PS2O), based model, instantiated tested illustrate ideas clearly. Experiments were conducted set 17 benchmark including both continuous discrete cases. The results demonstrate remarkable PS2O all chosen functions when compared several successful evolutionary algorithms.