edaReplace: Replacement Methods

In copulaedas: Estimation of Distribution Algorithms Based on Copulas

Description

Methods for the edaReplace generic function.

Usage

edaReplaceComplete(eda, gen, pop, popEval, sampledPop, sampledEval)
edaReplaceRTR(eda, gen, pop, popEval, sampledPop, sampledEval)

Arguments

eda	EDA instance.
gen	Generation.
pop	Matrix with one row for each solution in the population.
popEval	Vector with the evaluation of each solution in pop.
sampledPop	Matrix with one row for each solution sampled in the current generation.
sampledEval	Vector with the evaluation of the candidate solutions in sampledPop.

Details

Replacement methods combine the candidate solutions sampled in the current generation with the candidate solutions from the population of the previous generation. The following replacement methods are implemented.

edaReplaceComplete

The population sampled in the current generation completely replaces the population of the previous generation. This is the default method of the edaReplace generic function.

edaReplaceRTR

Restricted Tournament Replacement is a niching method that can be used to promote the preservation of alternative candidate solutions. See (Pelikan 2005) for a pseudocode of the algorithm implemented here. The parameter windowSize specifies the window size (default value: min(ncol(pop), nrow(pop) / 2)).

Value

A list with the following components.

pop	Matrix with one row for each solution in the new population.
popEval	Vector with the evaluation of each solution in pop.

References

Gonzalez-Fernandez Y, Soto M (2014). copulaedas: An R Package for Estimation of Distribution Algorithms Based on Copulas. Journal of Statistical Software, 58(9), 1-34. http://www.jstatsoft.org/v58/i09/.

copulaedas documentation built on May 1, 2019, 10:24 p.m.