Select parents using NSGA-II rank and crowding distance

Ranks the population into non-dominated fronts, computes crowding distances, and performs binary tournament selection that prefers lower rank and, on ties, higher crowding distance.

Usage

selection_cpp(population, fitness, num_parents)

Arguments

population

Numeric matrix of individuals.

fitness

Numeric matrix of fitness scores (rows: individuals, columns: objectives, all minimised).

num_parents

Number of individuals to select.

Value

Numeric matrix of selected individuals.

References

Deb K, Pratap A, Agarwal S, Meyarivan T (2002). "A fast and elitist multiobjective genetic algorithm: NSGA-II." IEEE Transactions on Evolutionary Computation, 6(2), 182-197. doi:10.1109/4235.996017

See also

initialize_population_cpp(), evaluate_fitness_cpp(), crossover_cpp(), mutation_cpp(), replacement_cpp(), bioga_main_cpp()

Examples

genomic_data <- matrix(rnorm(100), nrow = 10, ncol = 10)
population <- BioGA::initialize_population_cpp(genomic_data, population_size = 5)
fitness <- BioGA::evaluate_fitness_cpp(genomic_data, population, c(1.0, 0.5))
BioGA::selection_cpp(population, fitness, num_parents = 2)
#>             [,1]      [,2]       [,3]      [,4]     [,5]      [,6]       [,7]
#> [1,]  1.09116354 0.0700649 -1.0061661 0.2912297 1.385545 1.4358605 -0.4794401
#> [2,] -0.04448748 0.6941418 -0.3148323 0.4334064 1.385545 0.6116994 -0.2336441
#>           [,8]       [,9]      [,10]
#> [1,] -1.585889 -0.3838501  0.9133362
#> [2,] -2.679716 -0.2075835 -0.7181257