Multi-dueling bandits and their application to online ranker evaluation

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

20 Citations (Scopus)

Abstract

New ranking algorithms are continually being developed and refined, necessitating the development of efficient methods for evaluating these rankers. Online ranker evaluation focuses on the challenge of efficiently determining, from implicit user feedback, which ranker out of a finite set of rankers is the best. Online ranker evaluation can be modeled by dueling ban- dits, a mathematical model for online learning under limited feedback from pairwise comparisons. Comparisons of pairs of rankers is performed by interleaving their result sets and examining which documents users click on. The dueling bandits model addresses the key issue of which pair of rankers to compare at each iteration, thereby providing a solution to the exploration-exploitation trade-off. Recently, methods for simultaneously comparing more than two rankers have been developed. However, the question of which rankers to compare at each iteration was left open. We address this question by proposing a generalization of the dueling bandits model that uses simultaneous comparisons of an unrestricted number of rankers. We evaluate our algorithm on synthetic data and several standard large-scale online ranker evaluation datasets. Our experimental results show that the algorithm yields orders of magnitude improvement in performance compared to stateof- the-art dueling bandit algorithms.
Original languageEnglish
Title of host publicationProceedings of the 25th ACM International Conference on Information and Knowledge Management
Number of pages6
PublisherAssociation for Computing Machinery
Publication date2016
Pages2161-2166
ISBN (Electronic)978-1-4503-4073-1
DOIs
Publication statusPublished - 2016
Event25th ACM International Conference on Information and Knowledge Management - Indianapolis, United States
Duration: 24 Oct 201628 Oct 2016
Conference number: 25

Conference

Conference25th ACM International Conference on Information and Knowledge Management
Number25
Country/TerritoryUnited States
CityIndianapolis
Period24/10/201628/10/2016
SeriesACM International Conference on Information and Knowledge Management

Cite this