Practical Contextual Bandits with Regression Oracles

Dylan J. Foster, Alekh Agarwal, Miroslav Dudik, Luo Haipeng, Robert E. Schapire

Research output: Chapter in Book/Report/Conference proceedingConference contribution

24 Scopus citations


A major challenge in contextual bandits is to design general-purpose algorithms that are both practically useful and theoretically well-founded. We present a new technique that has the empiri-cal and computational advantages of realizability- based approaches combined with the flexibility of agnostic methods. Our algorithms leverage the availability of a regression oracle for the value- function class, a more realistic and reasonable oracle than the classification oracles over policies typically assumed by agnostic methods. Our approach generalizes both UCB and LinUCB to far more expressive possible model classes and achieves low regret under certain distributional as-sumptions. In an extensive empirical evaluation, we find that our approach typically matches or outperforms both realizability-based and agnostic baselines.

Original languageEnglish (US)
Title of host publication35th International Conference on Machine Learning, ICML 2018
EditorsJennifer Dy, Andreas Krause
PublisherInternational Machine Learning Society (IMLS)
Number of pages36
ISBN (Electronic)9781510867963
StatePublished - 2018
Externally publishedYes
Event35th International Conference on Machine Learning, ICML 2018 - Stockholm, Sweden
Duration: Jul 10 2018Jul 15 2018

Publication series

Name35th International Conference on Machine Learning, ICML 2018


Other35th International Conference on Machine Learning, ICML 2018

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Human-Computer Interaction
  • Software


Dive into the research topics of 'Practical Contextual Bandits with Regression Oracles'. Together they form a unique fingerprint.

Cite this