ICML 2017poster168 citations

On Context-Dependent Clustering of Bandits

Claudio Gentile, Shuai Li, Purushottam Kar, Alexandros Karatzoglou, Giovanni Zappella, Evans Etrue

Abstract

We investigate a novel cluster-of-bandit algorithm CAB for collaborative recommendation tasks that implements the underlying feedback sharing mechanism by estimating user neighborhoods in a context-dependent manner. CAB makes sharp departures from the state of the art by incorporating collaborative effects into inference, as well as learning processes in a manner that seamlessly interleaves explore-exploit tradeoffs and collaborative steps. We prove regret bounds for CAB under various data-dependent assumptions which exhibit a crisp dependence on the expected number of clusters over the users, a natural measure of the statistical difficulty of the learning task. Experiments on production and real-world datasets show that CAB offers significantly increased prediction performance against a representative pool of state-of-the-art methods.

BibTeX
@InProceedings{pmlr-v70-gentile17a,
  title = 	 {On Context-Dependent Clustering of Bandits},
  author =       {Claudio Gentile and Shuai Li and Purushottam Kar and Alexandros Karatzoglou and Giovanni Zappella and Evans Etrue},
  booktitle = 	 {Proceedings of the 34th International Conference on Machine Learning},
  pages = 	 {1253--1262},
  year = 	 {2017},
  editor = 	 {Precup, Doina and Teh, Yee Whye},
  volume = 	 {70},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {06--11 Aug},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v70/gentile17a/gentile17a.pdf},
  url = 	 {https://proceedings.mlr.press/v70/gentile17a.html},
  abstract = 	 {We investigate a novel cluster-of-bandit algorithm CAB for collaborative recommendation tasks that implements the underlying feedback sharing mechanism by estimating user neighborhoods in a context-dependent manner. CAB makes sharp departures from the state of the art by incorporating collaborative effects into inference, as well as learning processes in a manner that seamlessly interleaves explore-exploit tradeoffs and collaborative steps. We prove regret bounds for CAB under various data-dependent assumptions which exhibit a crisp dependence on the expected number of clusters over the users, a natural measure of the statistical difficulty of the learning task. Experiments on production and real-world datasets show that CAB offers significantly increased prediction performance against a representative pool of state-of-the-art methods.}
}
On Context-Dependent Clustering of Bandits · ICML 2017