Algorithm selection via meta-learning and sample-based active testing

S.M. Abdulrahman, P. Brazdil, J.N. van Rijn, J. Vanschoren

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

10 Citations (Scopus)
2 Downloads (Pure)

Abstract

Identifying the best machine learning algorithm for a given problem continues to be an active area of research. In this paper we present a new method which exploits both meta-level information acquired in past experiments and active testing, an algorithm selection strategy. Active testing attempts to iteratively identify an algorithm whose performance will most likely exceed the performance of previously tried algorithms. The novel method described in this paper uses tests on smaller data sample to rank the most promising candidates, thus optimizing the schedule of experiments to be carried out. The experimental results show that this approach leads to considerably faster algorithm selection. Keywords: Algorithm selection, Meta-learning, Active testing, Algorithm Ranking
Original languageEnglish
Title of host publicationInternational Workshop on Meta-Learning and Algorithm Selection (MetaSel 2015, Porto, Portugal September 7, 2015; co-located with ECMLPKDD 2015)
EditorsJ. Vanschoren, P. Brazdil, C. Giraud-Carrier, L. Kotthoff
PublisherCEUR-WS.org
Pages55-66
Publication statusPublished - 2015

Publication series

NameCEUR Workshop Proceedings
Volume1455
ISSN (Print)1613-0073

Fingerprint

Dive into the research topics of 'Algorithm selection via meta-learning and sample-based active testing'. Together they form a unique fingerprint.

Cite this