TY - GEN
T1 - Avoiding the pitfalls of noisy fitness functions with genetic algorithms
AU - Larkin, Fiacc
AU - Ryan, Conor
PY - 2009
Y1 - 2009
N2 - We have examined the application of genetic Algorithms to noisy fitness functions and consider the accepted wisdom of sampling, or multiple evaluations of individuals, as a mechanism for identifying true performance. Given a large (> 10%) amount of noise, a standard GA of surprisingly modest population size outperforms a GA using sampling, when compared on fitness versus evaluations. We also document a detrimental phenomenon we term the Glass Ceiling, which is when individuals of high fitness become confused with individuals of perfect fitness by the GA. We pinpoint the precise conditions that create this effect.
AB - We have examined the application of genetic Algorithms to noisy fitness functions and consider the accepted wisdom of sampling, or multiple evaluations of individuals, as a mechanism for identifying true performance. Given a large (> 10%) amount of noise, a standard GA of surprisingly modest population size outperforms a GA using sampling, when compared on fitness versus evaluations. We also document a detrimental phenomenon we term the Glass Ceiling, which is when individuals of high fitness become confused with individuals of perfect fitness by the GA. We pinpoint the precise conditions that create this effect.
KW - Clones
KW - Genetic algorithms
KW - Noisy fitness functions
KW - Performance
KW - The glass ceiling
UR - http://www.scopus.com/inward/record.url?scp=72749124481&partnerID=8YFLogxK
U2 - 10.1145/1569901.1570204
DO - 10.1145/1569901.1570204
M3 - Conference contribution
AN - SCOPUS:72749124481
SN - 9781605583259
T3 - Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009
SP - 1861
EP - 1862
BT - Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009
T2 - 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009
Y2 - 8 July 2009 through 12 July 2009
ER -