Memetic Algorithms
The term meme was coined by Dawkins in 1976 in his book The Selfish Gene [7 ]. The sociological definition of a meme is the basic unit of cultural transmission or imitation. A meme is the social analog of genes for individuals. Universal Darwinism draws t
- PDF / 154,813 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 96 Downloads / 185 Views
19
The term meme was coined by Dawkins in 1976 in his book The Selfish Gene [7]. The sociological definition of a meme is the basic unit of cultural transmission or imitation. A meme is the social analog of genes for individuals. Universal Darwinism draws the analogy on the role of genes in genetic evolution to that of memes in a cultural evolutionary process [7]. The science of memetics [3] represents the minduniverse analog to genetics in cultural evolution, ranging the fields of anthropology, biology, cognition, psychology, sociology and sociobiology. This chapter is dedicated to memetic and cultural algorithms.
19.1 Introduction The meme is a unit of intellectual or cultural information that can pass from mind to mind, when people exchange ideas. As genes propagate in the gene pool via sperms or eggs, memes propagate in the meme pool by spreading from brain to brain via a process called imitation. Unlike genes, memes are typically adapted by the people who transmit them before being passed on, that is, meme is a lifetime learning procedure capable of generating refinement on individuals. Like genes that serve as building blocks in genetics, memes are building blocks of meaningful information that is transmissible and replicable. Memes can be thought of as schemata that are modified and passed on over a learning process. The concept of schemata being passable are just as behaviors or thoughts are passed on memes. The typical memetic algorithm uses an additional mechanism to modify schemata during an individual’s lifetime, taken as the period of evaluation from the point of view of GA, and that refinement can be passed on to an individual’s offspring. Memetic computation is a computational paradigm that encompasses the construction of a comprehensive set of memes. It involves the additional dimension of cultural evolution through memetic transmission, selection, replication, imitation, or © Springer International Publishing Switzerland 2016 K.-L. Du and M.N.S. Swamy, Search and Optimization by Metaheuristics, DOI 10.1007/978-3-319-41192-7_19
315
316
19 Memetic Algorithms
variation, in the context of problem-solving. Memetic computation is the hybridization of a population-based global search and the local improvement, which strikes a balance between exploration and exploitation of the search space. An important step in memetic computation is to identify a suitable memetic representation of the memotype. The memetic evolutionary process is primarily driven by imitation [3], which takes place during transmission of memes. Individuals make choices and imitate others who have obtained high payoffs in the previous rounds. For imitation, memetic selection decides whom to imitate, memetic transmission decides how to imitate, and memetic variation relates to what is imitated or assimilated. In memetic expression and assimilation, the focus is placed on the socio-types (which is the social expression of a meme, as analogous to the phenotype of a gene) instead of memotypes of the agents. The agent assimilates memes
Data Loading...
