rgp (version 0.4-1)

makeArchiveBasedParetoTournamentSearchHeuristic: Archive-based Pareto Tournament Search Heuristic for RGP

Description

The search-heuristic, i.e. the concrete GP search algorithm, is a modular component of RGP. makeArchiveBasedParetoTournamentSearchHeuristic creates a RGP search-heuristic that implements a archive-based Pareto tournament multi objective optimization algorithm (EMOA) that selects on three criteria: Individual fitness, individual complexity and individual age.

Usage

makeArchiveBasedParetoTournamentSearchHeuristic(archiveSize = 50, popTournamentSize = 5, archiveTournamentSize = 3, crossoverRate = 0.95, enableComplexityCriterion = TRUE, complexityMeasure = function(ind, fitness) fastFuncVisitationLength(ind), ndsSelectionFunction = nds_cd_selection)

Arguments

archiveSize
The number of individuals in the archive, defaults to 50.
popTournamentSize
The size of the Pareto tournaments for selecting individuals for reproduction from the population.
archiveTournamentSize
The size of the Pareto tournaments for selecting individuals for reproduction from the archive.
crossoverRate
The probabilty to do crossover with an archive member instead of mutation of an archive member.
enableComplexityCriterion
Whether to enable the complexity criterion in multi-criterial search heuristics.
complexityMeasure
The complexity measure, a function of signature function(ind, fitness) returning a single numeric value.
ndsSelectionFunction
The function to use for non-dominated sorting in Pareto GP selection. Defaults to nds_cd_selection.

Value

An RGP search heuristic.