Memetic algorithms

The term memetic algorithms (MAs) was introduced in the late 1980s to denote a family of metaheuristics that have as central theme the hybridization of different algorithmic approaches for a given problem. Special emphasis was given to the use of a population-based approach in which a set of cooperating and competing agents was engaged in periods of individual improvement of the solutions while they interact sporadically. Another main objective theme was to introduce problem and instance-dependent knowledge as a way of speeding up the search process. Initialty, hybridizations included Evolutionary Algorithms (EAs), Simulated Annealing and its variants, and Tabu Search. Today, a number of hybridizations inciude other metaheuristics as well as exact algorithms, in complete anytime memetic algorithms. These methods not only prove optimality, but can also deliver high-quality soiutions early on in the process. MAs exploit problem knowledge by incorporating preexisting heuristics, preprocessing data reduction rules, approximation and fixed-parameter tractable algorithms, local search techniques, specialized recombination operators, truncated exact methods, and so on. Also, an important factor is the use of adequate representations of the problem being tackled. This results in highly efficient optimization tools. We provide a brief abstracted overview of MA applications in combinatorial optimization in Section 27.3. We will finish with a brief summary of the current research trends in MAs, with special mention to those which we believe will play a major role in the near future