Benefits of a population: five mechanisms that advantage population-based algorithms
Benefits of a population: five mechanisms that advantage population-based algorithms
This paper identifies five distinct mechanisms by which a population-based algorithm might have an advantage over a solo-search algorithm in classical optimization. These mechanisms are illustrated through a number of toy problems. Simulations are presented comparing different search algorithms on these problems. The plausibility of these mechanisms occurring in classical optimization problems is discussed. The first mechanism we consider relies on putting together building blocks from different solutions. This is extended to include problems containing critical variables. The second mechanism is the result of focusing of the search caused by crossover. Also discussed in this context is strong focusing produced by averaging many solutions. The next mechanism to be examined is the ability of a population to act as a low-pass filter of the landscape, ignoring local distractions. The fourth mechanism is a population's ability to search different parts of the fitness landscape, thus hedging against bad luck in the initial position or the decisions it makes. The final mechanism is the opportunity of learning useful parameter values to balance exploration against exploitation.
500-517
Prugel-Bennett, A.
b107a151-1751-4d8b-b8db-2c395ac4e14e
August 2010
Prugel-Bennett, A.
b107a151-1751-4d8b-b8db-2c395ac4e14e
Prugel-Bennett, A.
(2010)
Benefits of a population: five mechanisms that advantage population-based algorithms.
IEEE Transactions on Evolutionary Computation, 14 (4), .
(doi:10.1109/TEVC.2009.2039139).
Abstract
This paper identifies five distinct mechanisms by which a population-based algorithm might have an advantage over a solo-search algorithm in classical optimization. These mechanisms are illustrated through a number of toy problems. Simulations are presented comparing different search algorithms on these problems. The plausibility of these mechanisms occurring in classical optimization problems is discussed. The first mechanism we consider relies on putting together building blocks from different solutions. This is extended to include problems containing critical variables. The second mechanism is the result of focusing of the search caused by crossover. Also discussed in this context is strong focusing produced by averaging many solutions. The next mechanism to be examined is the ability of a population to act as a low-pass filter of the landscape, ignoring local distractions. The fourth mechanism is a population's ability to search different parts of the fitness landscape, thus hedging against bad luck in the initial position or the decisions it makes. The final mechanism is the opportunity of learning useful parameter values to balance exploration against exploitation.
Text
paper1.pdf
- Accepted Manuscript
More information
e-pub ahead of print date: 29 March 2010
Published date: August 2010
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 270918
URI: http://eprints.soton.ac.uk/id/eprint/270918
PURE UUID: 2d82ef47-5b0d-4471-bdee-354a838613b0
Catalogue record
Date deposited: 23 Apr 2010 14:13
Last modified: 14 Mar 2024 09:18
Export record
Altmetrics
Contributors
Author:
A. Prugel-Bennett
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics