Abstract
The majority of online users do not engage actively with what they are offered: they mostly use few items, give feedback on even fewer. Additionally, in many cases, the only feedback available about the item is positive feedback. These issues are well-known in the area of personalized recommendation and there have been many attempts to develop recommendation algorithms based on data consisting of only positive feedback. Most such state-of-the-art recommendation methods use convex loss functions, and either interpret non-interactivity with an item as negative feedback or ignore such entries altogether, none of which in principal reflects the reality. In this work, we provide reasons to motivate the usage of a non-convex loss in implicit feedback scenario to deal with unlabelled data, and devise an algorithm to minimize the proposed loss in collaborative setting. We analyse the effects of the proposed loss both qualitatively and quantitatively on a benchmark public dataset.
G. Singh was an intern at Yahoo at the time of the work.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Symmetric is defined as: \(f(x)+f(-x)=const.\), like in [5].
References
Baeza-Yates, R., Ribeiro-Neto, B., et al.: Modern Information Retrieval, vol. 463. ACM press, New York (1999)
Bouchard, G., Yin, D., Guo, S.: Convex collective matrix factorization. In: AISTATS, vol. 31, pp. 144–152 (2013)
Cai, J.-F., Candès, E.J., Shen, Z.: A singular value thresholding algorithm for matrix completion. SIAM J. Optim. 20(4), 1956–1982 (2010)
DeCoste, D.: Collaborative prediction using ensembles of maximum margin matrix factorizations. In: Proceedings of the 23rd International Conference on Machine Learning, pp. 249–256. ACM (2006)
du Plessis, M.C., Niu, G., Sugiyama, M.: Analysis of learning from positive and unlabeled data. In: Advances in Neural Information Processing Systems, pp. 703–711 (2014)
Guan, N., Tao, D., Luo, Z., Yuan, B.: Online nonnegative matrix factorization with robust stochastic approximation. IEEE Trans. Neural Netw. Learn. Syst. 23(7), 1087–1099 (2012)
Harper, F.M., Konstan, J.A.: The movielens datasets: history and context. ACM Trans. Interact. Intell. Syst. 5(4), 19:1–19:19 (2015)
Hu, Y., Koren, Y., Volinsky, C.: Collaborative filtering for implicit feedback datasets. In: 2008 Eighth IEEE International Conference on Data Mining, pp. 263–272. IEEE (2008)
Lee, D.D., Seung, H.S.: Algorithms for non-negative matrix factorization. In: Advances in Neural Information Processing Systems, pp. 556–562 (2000)
Lefevre, A., Bach, F., Févotte, C.: Online algorithms for nonnegative matrix factorization with the Itakura-Saito divergence. In: 2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), pp. 313–316. IEEE (2011)
Lim, D., McAuley, J., Lanckriet, G.: Top-n recommendation with missing implicit feedback. In: Proceedings of the 9th ACM Conference on Recommender Systems, pp. 309–312. ACM (2015)
Pan, R., Zhou, Y., Cao, B., Liu, N.N., Lukose, R., Scholz, M., Yang, Q.: One-class collaborative filtering. In: 2008 Eighth IEEE International Conference on Data Mining, pp. 502–511. IEEE (2008)
Rendle, S., Freudenthaler, C., Gantner, Z., Schmidt-Thieme, L.: BPR: Bayesian personalized ranking from implicit feedback. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, pp. 452–461. AUAI Press (2009)
Schnabel, T., Swaminathan, A., Singh, A., Chandak, N., Joachims, T.: Recommendations as treatments: debiasing learning and evaluation. arXiv preprint arXiv:1602.05352 (2016)
Singh, A.P., Gordon, G.J.: Relational learning via collective matrix factorization. In: Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2008, pp. 650–658. ACM, New York (2008)
Srebro, N., Rennie, J., Jaakkola, T.S.: Maximum-margin matrix factorization. In: Advances in Neural Information Processing Systems, pp. 1329–1336 (2004)
Xu, W., Liu, X., Gong, Y.: Document clustering based on non-negative matrix factorization. In: Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 267–273. ACM (2003)
Acknowledgement
GS would like to acknowledge support from the FRE Program at Yahoo!
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Singh, G., Mitrović, S. (2018). Benefits of Using Symmetric Loss in Recommender Systems. In: Pasi, G., Piwowarski, B., Azzopardi, L., Hanbury, A. (eds) Advances in Information Retrieval. ECIR 2018. Lecture Notes in Computer Science(), vol 10772. Springer, Cham. https://doi.org/10.1007/978-3-319-76941-7_26
Download citation
DOI: https://doi.org/10.1007/978-3-319-76941-7_26
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-76940-0
Online ISBN: 978-3-319-76941-7
eBook Packages: Computer ScienceComputer Science (R0)