Multi-criteria meta-parameter tuning for mono-objective stochastic metaheuristics Export

@inproceedings{citeulike:4214319, abstract = {One of the main difficulties of applying a stochastic metaheuristics to an optimization problem is to choose the best parameter setting. Our work suggest that this can be considered as a bi-objective problem, where one try to optimize both speed and precision of the underlying algorithm. Moreover, this objective function is perturbated by noise, due to the stochastic nature of the metaheuristics, thus necessiting appropriate estimation of the measure bias. In this article, we propose to use bi-objective metaheuristic along with a simple method of noise estimation to find the Pareto front of the bests sets of parameters. The method has the advantages of: aggregating the several parameters of the studied metaheuristic into a single one, permitting to study their relative influence on its behavior and comparing several metaheuristics.}, author = {Dr\'{e}o, Johann}, booktitle = {2nd International Conference on Metaheuristics and Nature Inspired Computing}, citeulike-article-id = {4214319}, day = {30}, keywords = {algorithm-engineering, evolutionary-computation, metaheuristic, parameter-setting, performance-assessment, stochastic-local-search}, location = {Hammamet, Tunisia}, month = {October}, posted-at = {2009-03-24 13:13:58}, priority = {0}, title = {Multi-criteria meta-parameter tuning for mono-objective stochastic metaheuristics}, year = {2008} }

TY - CONF ID - citeulike:4214319 L3 - citeulike-article-id:4214319 N2 - One of the main difficulties of applying a stochastic metaheuristics to an optimization problem is to choose the best parameter setting. Our work suggest that this can be considered as a bi-objective problem, where one try to optimize both speed and precision of the underlying algorithm. Moreover, this objective function is perturbated by noise, due to the stochastic nature of the metaheuristics, thus necessiting appropriate estimation of the measure bias. In this article, we propose to use bi-objective metaheuristic along with a simple method of noise estimation to find the Pareto front of the bests sets of parameters. The method has the advantages of: aggregating the several parameters of the studied metaheuristic into a single one, permitting to study their relative influence on its behavior and comparing several metaheuristics. CY - Hammamet, Tunisia TI - Multi-criteria meta-parameter tuning for mono-objective stochastic metaheuristics T2 - 2nd International Conference on Metaheuristics and Nature Inspired Computing KW - algorithm-engineering KW - evolutionary-computation KW - metaheuristic KW - parameter-setting KW - performance-assessment KW - stochastic-local-search AU - Dréo, Johann PY - 2008/10/30/ ER -