Implementation of the NSGA-II EMOA algorithm by Deb.
The NSGA-II merges the current population and the generated offspring and reduces it by means of the following procedure: It first applies the non dominated sorting algorithm to obtain the nondominated fronts. Starting with the first front, it fills the new population until the i-th front does not fit. It then applies the secondary crowding distance criterion to select the missing individuals from the i-th front.
nsga2( fitness.fun, n.objectives = NULL, n.dim = NULL, minimize = NULL, lower = NULL, upper = NULL, mu = 100L, lambda = mu, mutator = setup(mutPolynomial, eta = 25, p = 0.2, lower = lower, upper = upper), recombinator = setup(recSBX, eta = 15, p = 0.7, lower = lower, upper = upper), terminators = list(stopOnIters(100L)), ... )
fitness.fun: [function]
The fitness function.
n.objectives: [integer(1)]
Number of objectives of obj.fun. Optional if obj.fun is a benchmark function from package smoof.
n.dim: [integer(1)]
Dimension of the decision space.
minimize: [logical(n.objectives)]
Logical vector with ith entry TRUE if the ith objective of fitness.fun
shall be minimized. If a single logical is passed, it is assumed to be valid for each objective.
lower: [numeric]
Vector of minimal values for each parameter of the decision space in case of float or permutation encoding. Optional if obj.fun is a benchmark function from package smoof.
upper: [numeric]
Vector of maximal values for each parameter of the decision space in case of float or permutation encoding. Optional if obj.fun is a benchmark function from package smoof.
mu: [integer(1)]
Number of individuals in the population. Default is 100.
lambda: [integer(1)]
Offspring size, i.e., number of individuals generated by variation operators in each iteration. Default is 100.
mutator: [ecr_mutator]
Mutation operator of type ecr_mutator.
recombinator: [ecr_recombinator]
Recombination operator of type ecr_recombinator.
terminators: [list]
List of stopping conditions of type ecr_terminator . Default is to stop after 100 iterations.
...: [any]
Further arguments passed down to fitness function.
[ecr_multi_objective_result]
This is a pure R implementation of the NSGA-II algorithm. It hides the regular ecr interface and offers a more R like interface while still being quite adaptable.
Deb, K., Pratap, A., and Agarwal, S. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, 6 (8) (2002), 182-197.