updt_restricted {MOEADr}

Restricted Neighborhood Replacement Update for MOEA/D

Description

Population update using the restricted neighborhood replacement method for the MOEADr package.

Usage

updt_restricted(update, X, Xt, Y, Yt, V, Vt, sel.indx, B, ...)

Arguments

update

List containing the population update parameters. See Section ⁠Update Strategies⁠ of the moead() documentation for details. update must contain a field update$nr, a positive integer that determines the maximum number of copies of each candidate solution.

X

Matrix of candidate solutions

Xt

Matrix of incumbent solutions

Y

Matrix of objective function values of X

Yt

Matrix of objective function values of Xt

V

List object containing information about the constraint violations of the candidate solutions, generated by evaluate_population()

Vt

List object containing information about the constraint violations of the incumbent solutions, generated by evaluate_population()

sel.indx

matrix of selection indices, generated by order_neighborhood()

B

Neighborhood matrix, generated by define_neighborhood().

...

other parameters (included for compatibility with generic call)

Details

The restricted neighborhood replacement method behaves like the "standard" replacement method, except that each individual can only be selected up to nr times. After this limit has been reached, the next best individual in the same neighborhood is selected.

This update routine is intended to be used internally by the main moead() function, and should not be called directly by the user.

Value

List object containing the update population matrix (X), and its corresponding matrix of objective function values (Y) and constraint value list (V).

References

F. Campelo, L.S. Batista, C. Aranha (2020): The MOEADr Package: A Component-Based Framework for Multiobjective Evolutionary Algorithms Based on Decomposition. Journal of Statistical Software doi:10.18637/jss.v092.i06