Lexicographic Selection Function in ECR

I want to use a lexicographic selection function for a multi-objective genetic algorithm.

I could implement this myself but I was wondering if the default selTournament function in the ECR (Evolutionary Computing in R) library already uses lexicographic selection.

In the documentation, it simply states that it takes a matrix of fitness value(s), k for tournament size, and n.select for the number of candidates to return. It states that k candidates are selected at random from the population and the "best" is selected. This is repeated n.select times until it has formed the mating pool.

It does not state how this selection of "best" takes place in detail. For single fitness values, I imagine it would select the max out of those. However, as the documentation states that it accepts multiple fitness values per candidate, it does not state how this is handled: It could be lexicographically, through dominated hypervolume, crowding distance, aggregation with or without weighting, etc.

I cannot find any more information on the function. There is a selTournamentMO function in mosmafs, but this explicitly states that it uses dominated hypervolume or crowding as its criteria for selection, which is not what I need.

If the selTournament function in ECR uses lexicographic selection, then I can simply use this. Does anyone know how could I find out which methodology it employs?

Thank you.

ECR documentation:

mosmafs::selTournamentMO documentation:

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.