Development of the tournament selection-based genetic algorithms with the aid of a new selection method

2014 ◽  
Author(s):  
Chang-Wook Han
Keyword(s):  
Genetic Algorithms ◽  
Selection Method ◽  
Tournament Selection

scholarly journals Analysis of Selection Methods Used in Genetic Algorithms

2021 ◽  
Vol 4 ◽  
pp. 29-43
Author(s):  
Nataliya Gulayeva ◽  
Artem Ustilov
Keyword(s):  
Growth Rate ◽  
Genetic Algorithms ◽  
Selection Pressure ◽  
Fitness Function ◽  
Selection Method ◽  
Initial Population ◽  
Reproduction Rate ◽  
Selection Methods ◽  
Selection Scheme ◽  
Fitness Value

This paper offers a comprehensive review of selection methods used in the generational genetic algorithms.Firstly, a brief description of the following selection methods is presented: fitness proportionate selection methods including roulette-wheel selection (RWS) and its modifications, stochastic remainder selection with replacement (SRSWR), remainder stochastic independent selection (RSIS), and stochastic universal selection (SUS); ranking selection methods including linear and nonlinear rankings; tournament selection methods including deterministic and stochastic tournaments as well as tournaments with and without replacement; elitist and truncation selection methods; fitness uniform selection scheme (FUSS).Second, basic theoretical statements on selection method properties are given. Particularly, the selection noise, selection pressure, growth rate, reproduction rate, and computational complexity are considered. To illustrate selection method properties, numerous runs of genetic algorithms using the only selection method and no other genetic operator are conducted, and numerical characteristics of analyzed properties are computed. Specifically, to estimate the selection pressure, the takeover time and selection intensity are computed; to estimate the growth rate, the ratio of best individual copies in two consecutive populations is computed; to estimate the selection noise, the algorithm convergence speed is analyzed based on experiments carried out on a specific fitness function assigning the same fitness value to all individuals.Third, the effect of selection methods on the population fitness distribution is investigated. To do this, there are conducted genetic algorithm runs starting with a binomially distributed initial population. It is shown that most selection methods keep the distribution close to the original one providing an increased mean value of the distribution, while others (such as disruptive RWS, exponential ranking, truncation, and FUSS) change the distribution significantly. The obtained results are illustrated with the help of tables and histograms.

A Simple Approach to Lifetime Learning in Genetic Programming-Based Symbolic Regression

2014 ◽  
Vol 22 (2) ◽  
pp. 287-317 ◽  
Author(s):  
Raja Muhammad Atif Azad ◽  
Conor Ryan
Keyword(s):  
Genetic Programming ◽  
Local Search ◽  
Genetic Material ◽  
Selection Method ◽  
Symbolic Regression ◽  
Tree Size ◽  
Linear Scaling ◽  
Natural Evolution ◽  
Lifetime Experience ◽  
Tournament Selection

Genetic programming (GP) coarsely models natural evolution to evolve computer programs. Unlike in nature, where individuals can often improve their fitness through lifetime experience, the fitness of GP individuals generally does not change during their lifetime, and there is usually no opportunity to pass on acquired knowledge. This paper introduces the Chameleon system to address this discrepancy and augment GP with lifetime learning by adding a simple local search that operates by tuning the internal nodes of individuals. Although not the first attempt to combine local search with GP, its simplicity means that it is easy to understand and cheap to implement. A simple cache is added which leverages the local search to reduce the tuning cost to a small fraction of the expected cost, and we provide a theoretical upper limit on the maximum tuning expense given the average tree size of the population and show that this limit grows very conservatively as the average tree size of the population increases. We show that Chameleon uses available genetic material more efficiently by exploring more actively than with standard GP, and demonstrate that not only does Chameleon outperform standard GP (on both training and test data) over a number of symbolic regression type problems, it does so by producing smaller individuals and it works harmoniously with two other well-known extensions to GP, namely, linear scaling and a diversity-promoting tournament selection method.

Sign in / Sign up

Export Citation Format

Share Document