Success-History Based Position Adaptation in Co-operation of Biology Related Algorithms

Abstract

Previously, a meta-heuristic approach called Co-Operation of Biology Related Algorithms or COBRA based on a fuzzy logic controller for solving real-parameter optimization problems was introduced and described. COBRA’s basic idea consists in a cooperative work of well-known bio-inspired algorithms with similar schemes, while the fuzzy logic controller determines which bio-inspired algorithms should be included in the co-operative work at a given moment for solving optimization problems using the COBRA approach. COBRA’s performance has been evaluated on a set of test functions and its workability demonstrated. However, COBRA’s search efficiency depends significantly on its ability to keep the balance between exploration and exploitation when solving complex multimodal problems. In this study, a new technique for generating potential solutions in biology-inspired algorithms is proposed. This technique uses a historical memory of successful positions found by individuals to guide them in different directions and thus to improve the exploration and exploitation abilities. The proposed method was applied to the components of the COBRA approach. The modified meta-heuristic as well as its original variant and components (with and without the proposed modification), were evaluated on a set of various well-known test functions. The obtained experimental results are presented and compared. It was established that the fuzzy-controlled COBRA with success-history based position adaptation allows better solutions with the same computational effort to be found. Thus, the usefulness of the proposed position adaptation technique was demonstrated.