Practical and Secure Federated Recommendation with Personalized Masks

TL;DR

This paper introduces FedMMF, a federated recommendation method using personalized masks that protect user privacy without compromising accuracy or efficiency, suitable for real-time personalized recommendations.

Contribution

The paper proposes a novel federated matrix factorization approach with personalized masks generated locally, enhancing privacy and efficiency without accuracy loss.

Findings

Outperforms existing methods on real-world datasets

Provides theoretical security guarantees for personalized masks

Maintains high recommendation accuracy with privacy protection

Abstract

Federated recommendation addresses the data silo and privacy problems altogether for recommender systems. Current federated recommender systems mainly utilize cryptographic or obfuscation methods to protect the original ratings from leakage. However, the former comes with extra communication and computation costs, and the latter damages model accuracy. Neither of them could simultaneously satisfy the real-time feedback and accurate personalization requirements of recommender systems. In this paper, we proposed federated masked matrix factorization (FedMMF) to protect the data privacy in federated recommender systems without sacrificing efficiency and effectiveness. In more details, we introduce the new idea of personalized mask generated only from local data and apply it in FedMMF. On the one hand, personalized mask offers protection for participants' private data without effectiveness loss. On the other hand, combined with the adaptive secure aggregation protocol, personalized mask could further improve efficiency. Theoretically, we provide security analysis for personalized mask. Empirically, we also show the superiority of the designed model on different real-world data sets.