Abstract
This paper addresses the on-line recommendation problem facing new users and new items; we assume that no information is available neither about users, nor about the items. The only source of information is a set of ratings given by users to some items. By on-line, we mean that the set of users, and the set of items, and the set of ratings is evolving along time and that at any moment, the recommendation system has to select items to recommend based on the currently available information, that is basically the sequence of past events. We also mean that each user comes with her preferences which may evolve along short and longer scales of time; so we have to continuously update their preferences. When the set of ratings is the only available source of information, the traditional approach is matrix factorization. In a decision making under uncertainty setting, actions should be selected to balance exploration with exploitation; this is best modeled as a bandit problem. Matrix factors provide a latent representation of users and items. These representations may then be used as contextual information by the bandit algorithm to select items. This last point is exactly the originality of this paper: the combination of matrix factorization and bandit algorithms to solve the on-line recommendation problem. Our work is driven by considering the recommendation problem as a feedback controlled loop. This leads to interactions between the representation learning, and the recommendation policy.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
\(\tilde{O}\) means O up to a logarithmic term on T.
References
Abbasi-yadkori, Y., Pal, D., Szepesvari, C.: Improved algorithms for linear stochastic bandits. In: Proceedings of NIPS, pp. 2312–2320 (2011)
Agarwal, D., Chen, B.-C., Elango, P., Motgi, N., Park, S.-T., Ramakrishnan, R., Roy, S., Zachariah, J.: Online models for content optimization. In: Proceedings of NIPS, pp. 17–24 (2008)
Auer, P., Cesa-Bianchi, N., Fischer, P.: Finite-time analysis of the multiarmed bandit problem. Mach. Learn. 47, 235–256 (2002)
Bennett, J., Lanning, S., Netflix, N.: The Netflix prize. In: KDD Cup and Workshop (2007)
Cesa-Bianchi, N., Lugosi, G.: Combinatorial bandits. J. Comput. Syst. Sci. 78(5), 1404–1422 (2012)
Chatterjee, S.: Matrix estimation by universal singular value thresholding. pre-print (2012). http://arxiv.org/abs/1212.1247
Dhanjal, C., Gaudel, R., Clémençon, S.: Collaborative filtering with localised ranking. In: Proceedings of AAAI (2015)
Dror, G., Koenigstein, N., Koren, Y., Weimer, M.: The Yahoo! music dataset and kdd-cup 2011. In: Proceedings of KDD Cup (2011)
Feldman, S.: Personalization with contextual bandits. http://engineering.richrelevance.com/author/sergey-feldman/
Kohli, P., Salek, M., Stoddard, G.: A fast bandit algorithm for recommendations to users with heterogeneous tastes. In: Proceedings of AAAI, pp. 1135–1141 (2013)
Koren, Y., Bell, R., Volinsky, C.: Matrix factorization techniques for recommender systems. Computer 42(8), 30–37 (2009)
Langford, J., Strehl, A., Wortman, J.: Exploration scavenging. In: Proceedings of ICML, pp. 528–535. Omnipress (2008)
Li, L., Chu, W., Langford, J., Schapire, R.E.: A contextual-bandit approach to personalized news article recommendation. In: Proceedings of WWW, pp. 661–670. ACM, New York (2010)
Li, L., Chu, W., Langford, J., Wang, X.: Unbiased offline evaluation of contextual-bandit-based news article recommendation algorithms. In: Proceedings of WSDM, pp. 297–306. ACM (2011)
Mary, J., Garivier, A., Li, L., Munos, R., Nicol, O., Ortner, R., Preux, P.: ICML exploration and exploitation 3 - new challenges (2012)
Shani, G., Heckerman, D., Brafman, R.I.: An MDP-based recommender system. J. Mach. Learn. Res. 6, 1265–1295 (2005)
Shivaswamy, P.K., Joachims, T.: Online learning with preference feedback. In: NIPS Workshop on Choice Models and Preference Learning (2011)
Walsh, T.J., Szita, I., Diuk, C., Littman, M.L.: Exploring compact reinforcement-learning representations with linear regression (2012). CoRR abs/1205.2606
Weston, J., Yee, H., Weiss, R.J.: Learning to rank recommendations with the k-order statistic loss. In: Proceedings of RecSys, pp. 245–248. ACM (2013)
White, J.M.: Bandit Algorithms for Website Optimization. O’Reilly, USA (2012)
Yue, Y., Hong, S.A., Guestrin, C.: Hierarchical exploration for accelerating contextual bandits. In: Proceedings of ICML, pp. 1895–1902 (2012)
Zhou, Y., Wilkinson, D., Schreiber, R., Pan, R.: Large-scale parallel collaborative filtering for the netflix prize. In: Fleischer, R., Xu, J. (eds.) AAIM 2008. LNCS, vol. 5034, pp. 337–348. Springer, Heidelberg (2008)
Acknowledgements
Authors acknowledge the support of INRIA, and the stimulating environment of the research group SequeL.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Mary, J., Gaudel, R., Preux, P. (2015). Bandits and Recommender Systems. In: Pardalos, P., Pavone, M., Farinella, G., Cutello, V. (eds) Machine Learning, Optimization, and Big Data. MOD 2015. Lecture Notes in Computer Science(), vol 9432. Springer, Cham. https://doi.org/10.1007/978-3-319-27926-8_29
Download citation
DOI: https://doi.org/10.1007/978-3-319-27926-8_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-27925-1
Online ISBN: 978-3-319-27926-8
eBook Packages: Computer ScienceComputer Science (R0)