The Price of Privacy in Untrusted Recommender Systems

Abstract : —Recent increase in online privacy concerns prompts the following question: can a recommender system be accurate if users do not entrust it with their private data? To answer this, we study the problem of learning item-clusters under local differential privacy, a powerful, formal notion of data privacy. We develop bounds on the sample-complexity of learning item-clusters from privatized user inputs. Significantly, our results identify a sample-complexity separation between learning in an information-rich and an information-scarce regime, thereby highlighting the interaction between privacy and the amount of information (ratings) available to each user. In the information-rich regime, where each user rates at least a constant fraction of items, a spectral clustering approach is shown to achieve a sample-complexity lower bound derived from a simple information-theoretic argument based on Fano's inequality. However, the information-scarce regime, where each user rates only a vanishing fraction of items, is found to require a fundamentally different approach both for lower bounds and algorithms. To this end, we develop new techniques for bounding mutual information under a notion of channel-mismatch. These techniques may be of broader interest, and we illustrate this by applying them to (i) learning based on 1-bit sketches, and (ii) adaptive learning. Finally, we propose a new algorithm, MaxSense, and show that it achieves optimal sample-complexity in the information-scarce regime.
Type de document :
Article dans une revue
IEEE Journal of Selected Topics in Signal Processing, IEEE, 2015, IEEE Journal of Topics in Signal Processing,, 9 (7), pp.1319 - 1331. 〈10.1109/JSTSP.2015.2423254〉
Liste complète des métadonnées

Littérature citée [17 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01226756
Contributeur : Laurent Massoulié <>
Soumis le : lundi 16 novembre 2015 - 10:20:35
Dernière modification le : jeudi 7 février 2019 - 17:34:15

Identifiants

Citation

Siddhartha Banerjee, Nidhi Hegde, Laurent Massoulié. The Price of Privacy in Untrusted Recommender Systems. IEEE Journal of Selected Topics in Signal Processing, IEEE, 2015, IEEE Journal of Topics in Signal Processing,, 9 (7), pp.1319 - 1331. 〈10.1109/JSTSP.2015.2423254〉. 〈hal-01226756〉

Partager

Métriques

Consultations de la notice

309