A Fast Bandit Algorithm for Recommendation to Users with Heterogenous Tastes

Pushmeet Kohli, Mahyar Salek, Greg Stoddard

AAAI 2013 pp. 1135-1141

doi:10.1609/AAAI.V27I1.8463 /aaai/2013/kohli2013aaai-fast/

Abstract

We study recommendation in scenarios where there's no prior information about the quality of content in the system. We present an online algorithm that continually optimizes recommendation relevance based on behavior of past users. Our method trades weaker theoretical guarantees in asymptotic performance than the state-of-the-art for stronger theoretical guarantees in the online setting. We test our algorithm on real-world data collected from previous recommender systems and show that our algorithm learns faster than existing methods and performs equally well in the long-run.

PDF AAAI Semantic Scholar

Cite

Text

Kohli et al. "A Fast Bandit Algorithm for Recommendation to Users with Heterogenous Tastes." AAAI Conference on Artificial Intelligence, 2013. doi:10.1609/AAAI.V27I1.8463

Markdown

[Kohli et al. "A Fast Bandit Algorithm for Recommendation to Users with Heterogenous Tastes." AAAI Conference on Artificial Intelligence, 2013.](https://mlanthology.org/aaai/2013/kohli2013aaai-fast/) doi:10.1609/AAAI.V27I1.8463

BibTeX

@inproceedings{kohli2013aaai-fast,
  title     = {{A Fast Bandit Algorithm for Recommendation to Users with Heterogenous Tastes}},
  author    = {Kohli, Pushmeet and Salek, Mahyar and Stoddard, Greg},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2013},
  pages     = {1135-1141},
  doi       = {10.1609/AAAI.V27I1.8463},
  url       = {https://mlanthology.org/aaai/2013/kohli2013aaai-fast/}
}