Ockham's Razor in memetic computing: Three stage optimal memetic exploration

摘要: Memetic computing is a subject in computer science which considers complex structures as the combination of simple agents, memes, whose evolutionary interactions lead to intelligent capable problem-solving. This paper focuses on memetic optimization algorithms and proposes counter-tendency approach for algorithmic design. Research field tends go direction improving existing by combining different methods or through formulation more complicated structures. Contrary this trend, we instead focus simplicity, proposing structurally algorithm with emphasis processing only one solution at time. The proposed algorithm, namely three stage optimal exploration, composed memes; first stochastic long search radius, second moderate radius third deterministic short radius. bottom-up operators means natural trial error logic, generates robust efficient optimizer, competing modern computationally expensive algorithms. suggestive fact that complexity can be unnecessary, if not detrimental, approaches are likely competitive here invoked an extension basing philosophical concept Ockham's Razor. An extensive experimental setup various test problems digital signal application presented. Numerical results show approach, despite its simplicity low computational cost displays very good performance several problems, sophisticated representing the-state-of-the-art intelligence optimization.