Article Preview
TopIntroduction
Evolutionary algorithms (EAs) were introduced by various researchers in the later part of 1960 (Fogel et al., 1996; Holland, 1975; Rechenberg, 1973). Memetic algorithms the term coined by Pablo Moscato (Moscato, 1989), actually covers a broad range of meta heuristics. It started to denote a family of meta-heuristics that tried to bring together concepts from tightly separated at that time families such as EAs and Simulated Annealing (SAs) (Kirkpatrick et al., 1983). They deviate from the traditional Evolutionary approaches as they explore all available knowledge about the problem under consideration. MAs are gaining huge success in giving practical solutions in variety of problem domains specifically Optimization problems.
The history of MA has its origin associated with the term ‘meme’ coined by Richard Dawkins (Dawkins, 1976) where he defines ‘meme’ is a part of cultural evolution which mimics a gene in genetic evolution. But, as is said, there are two sides to every coin, there are certain points where some techniques of hardcore mathematics do help. In a book titled “Where Darvin Goes Wrong” 2010 book by cognitive scientist Massimo Piattelli-Palmarini, and Jerry Fodor discusses certain key issue in EA. According to the original concept of MA, memes in form of information packets are passed from one human brain to another wherein they can be modified before passing on. Examples of meme given by Richard Dawkins are “tunes, ideas, catch-phrases, clothes fashions, ways of making pots or of building arches”. Another example of it can be the stories we tell to our children which we have heard from our parents which they have heard from their parents and so on. The contents can be modified from generation to generation and also, they can be modified for better. Genes are propagated as they are inherited from a parent to child which is not the case with memes. Memes inculcate a process of lifetime learning. As an example, the philosophy of Yoga is transmitted from generations to generations. When a person learns it, it becomes life time learning. A person can also enhance the process for better and some concepts which are not that interesting to that person can fade away. Memeticists argue that the memes most beneficial to their hosts will not necessarily thrive, rather, those memes that replicate the most effectively propagate. This is where there is a scope of introducing some mathematics. To improve meme locally, just as self-learning and observational skills are required, some mathematical skills too are most essential. Based on computational skill only, the meme can evolve or it may go absolutely wrong. For example, after observing fashion and without analyzing it in context with culture, weather, geographic and atmospheric conditions and cultural values, the evolution of meme could go absolutely wrong. Hence there is a strong need to provide bounded multidimensional integrated derivative outlining to evolution technique in frame of predictive optimization and method of elimination.