Paper Info
Title
Stochastic local search in continuous domains: questions to be answered when designing a novel algorithm
Abstract
Several population-based methods (with origins in the world of evolutionary strategies and estimation-of-distribution algorithms) for black-box optimization in continuous domains are surveyed in this article. The similarities and differences among them are emphasized and it is shown that they all can be described in a common framework of stochastic local search -- a class of methods previously defined mainly for combinatorial problems. Based on the lessons learned from the surveyed algorithms, a set of algorithm features (or, questions to be answered) is extracted. An algorithm designer can take advantage of these features and by deciding on each of them, she can construct a novel algorithm. A few examples in this direction are shown.
Year
DOI
Venue
2010
10.1145/1830761.1830830
GECCO (Companion)
Keywords
Field
DocType
evolutionary strategy,algorithm designer,common framework,algorithm feature,continuous domain,population-based method,estimation-of-distribution algorithm,novel algorithm,black-box optimization,combinatorial problem,stochastic local search,algorithm design,covariance matrix adaptation,cauchy distribution,probabilistic model,estimation of distribution algorithm,gaussian distribution,premature convergence
Population,Search algorithm,Estimation of distribution algorithm,Computer science,Cauchy distribution,Evolution strategy,Artificial intelligence,CMA-ES,Mathematical optimization,Premature convergence,Algorithm,Local search (optimization),Machine learning
Conference
Citations 
PageRank 
References 
1
0.38
21
Authors
1
Name
Order
Citations
PageRank
Petr Pošík121015.44