home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.genetic
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!cis.ohio-state.edu!udecc.engr.udayton.edu!blackbird.afit.af.mil!lmerkle
- From: lmerkle@afit.af.mil (Laurence D. Merkle)
- Subject: Re: GA versus Simulated Annealing
- Message-ID: <1993Jan24.230630.6734@afit.af.mil>
- Sender: news@afit.af.mil
- Nntp-Posting-Host: wb11.afit.af.mil
- Organization: Air Force Institute of Technology
- References: <1993Jan19.155159.16736@thunder.mcrcim.mcgill.edu>
- Date: Sun, 24 Jan 1993 23:06:30 GMT
- Lines: 29
-
- In article <1993Jan19.155159.16736@thunder.mcrcim.mcgill.edu> gblais@McRCIM.McGill.EDU (Gerard Blais) writes:
- >
- >I have been using a genetic algorithm to solve a minimization
- >problem. I performed the same minimization using the "Very Fast
- >Simulated Reannealing" algorithm created by Lester Ingber and
- >Bruce Rosen (I got it from a friend, but I believe it's available
- >from the net). In all cases I tried, the Annealing search was
- >orders of magnitude faster for converging to the global minimum
- >than the GA. My friend observed the same result for a different
- >minimization problem.
- >
-
- I'm interested in your results. I don't (yet) have any to offer
- along the same lines, but I have a few questions:
-
- What are your problems?
-
- Are they well behaved, or do they have lots of local minima?
-
- Are they large, or toy sized?
-
- Thanks!
-
- Larry
- --
- Laurence D. Merkle, CAPT, USAF | The opinions expressed are
- Student, Department of Electrical | my own, and do not necessarily
- and Computer Engineering | reflect those of AFIT, the USAF,
- Air Force Institute of Technology | or the US Government
-