🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: sci.math.num-analysis
Article #956

Re: function minimization references

#956
From: Richard Frost
Date: Thu, 17 Dec 1992 18:07
35 lines
1679 bytes
Gradient decent is fine if you can show there exist only a small
number of extrema in the search domain.

Here are some references which discuss alternative approaches.

SIMULATED ANNEALING:

J.R. Harland and P. Salamon "Simulated annealing: A review of the
thermodynamic approach" Nuclear Physics B (Proc. Suppl.) 5A (1988) pp. 109-115

G. Ruppeiner, et al "Ensemble approach to simulated annealing"
J. Phys. I vol. 1 (1991) 455-470

K.H. Hoffman, et al "Optimal ensemble size for parallel implementations
of simulated annealing" Appl. Math. Lett. vol. 3, no. 3 (1990) 53-56

SIMULATED RE-ANNEALING:

L. Ingber and B. Rosen "Genetic algorithms and very fast simulated
annealing: a comparison" Mathematical and Computer Modelling, 16(11)
1992, 87-100

L. Ingber "Very fast simulated re-annealing" Mathematical and Computer
Modelling, 12(8) 1989, 967-973

GENETIC ALGORITHMS:

K.A. De Jong "Genetic algorithms are NOT function optimizers" in
Foundations of Genetic Algorithms: Proceedings 24-29 July 1992, D.
Whitley (ed.) Morgan Kaufman

D.H. Ackley "An empirical study of bit vector function optimization" in
Genetic Algorithms and Simulated Annealing, L. Davis (ed.) Morgan
Kaufman (1987)
Technical Details
Message-ID: <1gqfl8INNe1b@network.ucsd.edu>
Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!network.ucsd.edu!stimpy!frost

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!network.ucsd.edu!stimpy!frost
From: frost@sdsc.edu (Richard Frost)
Newsgroups: sci.math.num-analysis
Subject: Re: function minimization references
Date: 17 Dec 1992 18:07:36 GMT
Organization: San Diego Supercomputer Center @UCSD.
Lines: 37
Distribution: world
Message-ID: <1gqfl8INNe1b@network.ucsd.edu>
Reply-To: frost@sdsc.edu
NNTP-Posting-Host: stimpy.sdsc.edu