Hybrid Evolutionary Algorithms: Methodologies, Architectures, and Reviews

Evolutionary computation has become an important problem solving methodology among many researchers. The population-based collective learning process, selfadaptation, and robustness are some of the key features of evolutionary algorithms when compared to other global optimization techniques. Even though evolutionary computation has been widely accepted for solving several important practical applications in engineering, business, commerce, etc., yet in practice sometimes they deliver only marginal performance. Inappropriate selection of various parameters, representation, etc. are frequently blamed. There is little reason to expect that one can find a uniformly best algorithm for solving all optimization problems. This is in accordance with the No Free Lunch theorem, which explains that for any algorithm, any elevated performance over one class of problems is exactly paid for in performance over another class. Evolutionary algorithm behavior is determined by the exploitation and exploration relationship kept throughout the run. All these clearly illustrates the need for hybrid evolutionary approaches where the main task is to optimize the performance of the direct evolutionary approach. Recently, hybridization of evolutionary algorithms is getting popular due to their capabilities in handling several real world problems involving complexity, noisy environment, imprecision, uncertainty, and vagueness. In this chapter, first we emphasize the need for hybrid evolutionary algorithms and then we illustrate the various possibilities for hybridization of an evolutionary algorithm and also present some of the generic hybrid evolutionary architectures that has evolved during the last couple of decades. We also provide a review of some of the interesting hybrid frameworks reported in the literature.