home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!enterpoop.mit.edu!mintaka.lcs.mit.edu!ai-lab!great-grain!mdlm
- From: mdlm@great-grain.ai.mit.edu (Michael De-la-Maza)
- Newsgroups: comp.ai.genetic
- Subject: Sigma truncation fitness function
- Date: 9 Jan 1993 21:53:50 GMT
- Organization: MIT AI Lab
- Lines: 30
- Sender: mdlm@great-grain (Michael De-la-Maza)
- Distribution: world
- Message-ID: <1inhheINNhr@life.ai.mit.edu>
- NNTP-Posting-Host: great-grain.ai.mit.edu
- Keywords: fitness function
-
- On p. 124 of Goldberg (1989), it says that the sigma truncation function
- is:
-
- f' = f - (<f> - c*sigma)
-
- ^^^
-
- On p. 62 of Michalewicz (1992) it says that the sigma truncation
- function is:
-
- f' = f + (<f> - c*sigma)
-
- ^^^
-
- Is it plus or minus? Goldberg cites Forrest (1985) as the original
- source, but I do not have a copy. Curiously enough, Michalewicz (1992)
- says that he is following Goldberg (1989).
-
- References
-
- Forrest (1985). Documentation for PRISONERS DILEMMA and NORMS programs
- that use the genetic algorithm. Unpublished manuscript. University of
- Michigan, Ann Arbor.
-
- Goldberg (1989). Genetic algorithms in search, optimization, and machine
- learning. Addison-Wesley.
-
- Michalewicz (1992). Genetic algorithms + data structures = evolution
- programs. Springer-Verlag.
-
-