DSpace at Cochin University >
Department of Computer Science >
Seminar Reports >
MTech 2010-2012 Batch >
Please use this identifier to cite or link to this item:
|Title: ||Evolutionary Programming|
|Authors: ||Savitha Sam Abraham|
|Keywords: ||Evolutionary Programming, Machine Learning, Mutation|
|Issue Date: ||22-Jun-2011|
|Abstract: ||Evolutionary programming was originally proposed in 1962 by Lawrence J Fogel as an alternative method for generating machine intelligence. Intelligent behavior requires the composite ability to predict ones environment coupled with a translation of the predictions into a suitable response in light of a given goal. Contemporary evolutionary programming is also used for optimization problems. Large dimensionality, non linearity of functions etc make some optimization problems complex where evolutionary programming can be used.
The first step in evolutionary programming is initialization where the first generation of individuals or candidate solutions is created. Then each individual is evaluated using a fitness function after which parent selection is done. Parents are mutated to generate offspring. Each offspring is then evaluated and then survivor selection is done. The different survivor selection techniques and types of evolutionary programming are also discussed. One feature of evolutionary programming is that it uses only mutation operators and does not use recombination. Depending on the problem, different mutation distributions like Gaussian, Cauchy distributions are used|
|Appears in Collections:||MTech 2010-2012 Batch|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.