Memetic programming with adaptive local search using tree data structures

Abstract

Meta-heuristics are general frameworks of heuristics methods for solving combinatorial optimization problems, where exploring the exact solutions for these problems becomes very hard due to some limitations like extremely large running time. In this paper, new local searches over tree space are defined. Using these local searches, various meta-heuristics can be generalized to deal with tree data structures to introduce a more general framework of meta-heuristics called Meta-Heuristics Programming (MHP) as general machine learning tools. As an alternative to Genetic Programming (GP) algorithm, Memetic Programming (MP) algorithm is proposed as a new outcome of the MHP framework. The efficiency of the proposed MP Algorithm is examined through comparative numerical experiments.