Skip to Main content Skip to Navigation
Conference papers

Illustrating the trade-off between time, quality, and success probability in heuristic search

Abstract : Benchmarking aims to investigate the performance of one or several algorithms for a set of reference problems by empirical means. An important motivation for benchmarking is the generation of insight that can be leveraged for designing more efficient solvers, for selecting a best algorithm, and/or for choosing a suitable instantiation of a parametrized algorithm. An important component of benchmarking is its design of experiment (DoE), which comprises the selection of the problems, the algorithms, the computational budget, etc., but also the performance indicators by which the data is evaluated. The DoE very strongly depends on the question that the user aims to answer. Flexible benchmarking environments that can easily adopt to users' needs are therefore in high demand. With the objective to provide such a flexible benchmarking environment, the recently released IOHprofiler not only allows the user to choose the sets of benchmark problems and reference algorithms, but provides in addition a highly interactive, versatile performance evaluation. However, it still lacks a few important performance indicators that are relevant to practitioners and/or theoreticians. In this discussion paper we focus on one particular aspect, the probability that a considered algorithm reaches a certain target value within a given time budget. We thereby suggest to extend the classically regarded fixed-target and fixed-budget analyses by a fixed-probability measure. Fixed-probability curves are estimated using Pareto layers of (target, budget) pairs that can be realized by the algorithm with the required certainty. We also provide a first implementation towards a fixed-probability module within the IOHprofiler environment.
Document type :
Conference papers
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://hal.sorbonne-universite.fr/hal-02179607
Contributor : Carola Doerr <>
Submitted on : Tuesday, January 14, 2020 - 3:40:56 PM
Last modification on : Friday, January 17, 2020 - 4:46:28 PM
Document(s) archivé(s) le : Wednesday, April 15, 2020 - 8:06:19 PM

File

bl Ignashov.Buzdalova.Buzdalov...
Files produced by the author(s)

Identifiers

Citation

Ivan Ignashov, Arina Buzdalova, Maxim Buzdalov, Carola Doerr. Illustrating the trade-off between time, quality, and success probability in heuristic search. Genetic and Evolutionary Computation Conference, Companion Material, Jul 2019, Prague, Czech Republic. pp.1807-1812, ⟨10.1145/3319619.3326895⟩. ⟨hal-02179607⟩

Share

Metrics

Record views

72

Files downloads

81