Free cookie consent management tool by TermsFeed Policy Generator

source: branches/OKBJavaConnector/ECJClient/src/ec/es/README @ 13067

Last change on this file since 13067 was 6152, checked in by bfarka, 14 years ago

added ecj and custom statistics to communicate with the okb services #1441

File size: 1.3 KB
Line 
1This package implements the (mu+lambda) and (mu,lambda) evolution strategies.
2Use this package in conjunction with the 'simple' package (that is,
3generational evolution).  The provided es.params file shows some more
4information on how to do this.
5
6These techniques are implemented with custom Breeders in combination with a
7special SelectionMethod which handles the dirty work of truncation selection.
8The SelectionMethod is:
9
10  ec.es.ESSelection
11
12This method should only be used with the Breeders in this package.  You can
13place ESSelection anywhere in your pipeline, including multiple times if you
14like.  It will return the mu best individuals in the population, one at a time,
15each time it's pulsed.  Individual ESSelection methods in various parts of
16your pipeline will return the same individual.
17
18The (mu,lambda) evolution strategy is implemented with the Breeder
19
20  ec.es.MuCommaLambdaBreeder
21
22This breeder requires you to specify a 'mu' value and a 'lambda' value for
23each subpopulation.  You need to provide a 'lambda' value instead of just
24the subpopulation size because the subpopulation size is used solely to set the
25INITIAL size of the subpopulation. 
26
27The (mu+lambda) evolution strategy has exactly the same options, and is
28implemented in a subclass:
29
30  ec.es.MuPlusLambdaBreeder
Note: See TracBrowser for help on using the repository browser.