Proceedings of the Conference on Recommender Syst
Proceedings of the Conference on Recommender Systems (RecSys) Year 2021 Peer-reviewed
BibTeX PDF
Web Security · Privacy

Stronger Privacy for Federated Collaborative Filtering With Implicit Feedback

Lorenzo Minto Moritz Haller Hamed Haddadi Benjamin Livshits
2021
Publication year
RecSys
Venue
Peer-reviewed
Type
§ Abstract

Problem

Recommender systems are commonly trained on centrally-collected user interaction data like views or clicks. This practice however raises serious privacy concerns regarding the recommender’s collection and handling of potentially sensitive data.

Approach

Several privacyaware recommender systems have been proposed in recent literature, but comparatively little attention has been given to systems at the intersection of implicit feedback and privacy. To address this shortcoming, we propose a practical federated recommender system for implicit data under user-level local differential privacy (LDP).

Results

The privacy-utility trade-off is controlled by parameters 𝜖 and 𝑘, regulating the per-update privacy budget and the number of 𝜖-LDP gradient updates sent by each user, respectively. To further protect the user’s privacy, we introduce a proxy network to reduce the fingerprinting surface by anonymizing and shuffling the reports before forwarding them to the recommender. We empirically demonstrate the effectiveness of our framework on the MovieLens dataset, achieving up to Hit Ratio with K=10 (HR@10) 0.68 on 50,000 users with 5,000 items. Even on the full dataset, we show that it is possible to achieve reasonable utility with HR@10>0.5 without compromising user privacy.

Cite this paper — BibTeX 
@inproceedings{livshits-recsys21,
  title = "Stronger Privacy for Federated Collaborative Filtering With
Implicit Feedback",
  author = "Lorenzo Minto and Moritz Haller and Hamed Haddadi and Benjamin Livshits",
  year = "2021",
  month = sep,
  booktitle = {Proceedings of the  Conference on Recommender Systems (RecSys)}
}
Copied