Upper Bounds on the Running Time of the Univariate Marginal Distribution Algorithm on OneMax

The Univariate Marginal Distribution Algorithm (UMDA) is a randomized search heuristic that builds a stochastic model of the underlying optimization problem by repeatedly sampling λ solutions and adjusting the model according to the best μ samples. We present a running time analysis of the UMDA on t...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Algorithmica Ročník 81; číslo 2; s. 632 - 667
Hlavní autor: Witt, Carsten
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York Springer US 15.02.2019
Springer Nature B.V
Témata:
ISSN:0178-4617, 1432-0541
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The Univariate Marginal Distribution Algorithm (UMDA) is a randomized search heuristic that builds a stochastic model of the underlying optimization problem by repeatedly sampling λ solutions and adjusting the model according to the best μ samples. We present a running time analysis of the UMDA on the classical OneMax benchmark function for wide ranges of the parameters μ and λ . If μ ≥ c log n for some constant  c > 0 and λ = ( 1 + Θ ( 1 ) ) μ , we obtain a general bound O ( μ n ) on the expected running time. This bound crucially assumes that all marginal probabilities of the algorithm are confined to the interval [ 1 / n , 1 - 1 / n ] . If μ ≥ c ′ n log n for a constant c ′ > 0 and λ = ( 1 + Θ ( 1 ) ) μ , the behavior of the algorithm changes and the bound on the expected running time becomes O ( μ n ) , which typically holds even if the borders on the marginal probabilities are omitted. The results supplement the recently derived lower bound Ω ( μ n + n log n ) by Krejca and Witt (Proceedings of FOGA 2017, ACM Press, New York, pp 65–79, 2017 ) and turn out to be tight for the two very different choices μ = c log n and μ = c ′ n log n . They also improve the previously best known upper bound O ( n log n log log n ) by Dang and Lehre (Proceedings of GECCO ’15, ACM Press, New York, pp 513–518, 2015 ) that was established for μ = c log n and λ = ( 1 + Θ ( 1 ) ) μ .
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0178-4617
1432-0541
DOI:10.1007/s00453-018-0463-0