NAME

Algorithm::Evolutionary::Fitness::P_Peaks - P Peaks problem generator

SYNOPSIS

my $number_of_peaks = 100;
my $number_of_bits = 32;
my $p_peaks = Algorithm::Evolutionary::Fitness::P_Peaks->new( $number_of_peaks, $number_of_bits );

DESCRIPTION

P_Peaks fitness function; optimizes the distance to the closest in a series of peaks. The P-Peaks problem was proposed by Kennedy and Spears in

 @conference{kennedy1998matching,
 title={{Matching algorithms to problems: an experimental test of the particle swarm and some genetic algorithms on the multimodal problem generator}},
 author={Kennedy, J. and Spears, W.M.},
 booktitle={Evolutionary Computation Proceedings, 1998. IEEE World Congress on Computational Intelligence., The 1998 IEEE International Conference on},
 pages={78--83},
 isbn={0780348699},
 year={1998},
 organization={IEEE}
}

And the optimum is 1.0. By default, result is cached, so be careful when working with long strings and/or big populations

METHODS

new( $peaks, $bits )

Creates a new instance of the problem, with the said number of bits and peaks.

random_string()

Returns random string in the same style than the peaks. Useful for testing.

_really_apply( $string )

Applies the instantiated problem to a chromosome. Intended for internal use.

p_peaks( $string )

Returns the distance to the closest bitstring

Copyright

This file is released under the GPL. See the LICENSE file included in this distribution,
or go to http://www.fsf.org/licenses/gpl.txt

CVS Info: $Date: 2011/04/08 06:43:05 $ 
$Header: /cvsroot/opeal/Algorithm-Evolutionary/lib/Algorithm/Evolutionary/Fitness/P_Peaks.pm,v 3.1 2011/04/08 06:43:05 jmerelo Exp $ 
$Author: jmerelo $ 
$Revision: 3.1 $
$Name $