Federated Unlearning for On-Device Recommendation

TL;DR

This paper introduces FRU, an efficient federated unlearning method for recommendation systems that enables users to erase their data contributions, ensuring privacy compliance and robustness against malicious attacks.

Contribution

The paper proposes a novel unlearning approach for federated recommendation systems, including a log-based rollback mechanism and techniques to reduce storage of historical updates.

Findings

FRU effectively removes user influence from federated recommenders.

The negative sampling and importance-based update selection improve efficiency.

Experiments demonstrate FRU's effectiveness and efficiency on real datasets.

Abstract

The increasing data privacy concerns in recommendation systems have made federated recommendations (FedRecs) attract more and more attention. Existing FedRecs mainly focus on how to effectively and securely learn personal interests and preferences from their on-device interaction data. Still, none of them considers how to efficiently erase a user's contribution to the federated training process. We argue that such a dual setting is necessary. First, from the privacy protection perspective, ``the right to be forgotten'' requires that users have the right to withdraw their data contributions. Without the reversible ability, FedRecs risk breaking data protection regulations. On the other hand, enabling a FedRec to forget specific users can improve its robustness and resistance to malicious clients' attacks. To support user unlearning in FedRecs, we propose an efficient unlearning method FRU (Federated Recommendation Unlearning), inspired by the log-based rollback mechanism of transactions in database management systems. It removes a user's contribution by rolling back and calibrating the historical parameter updates and then uses these updates to speed up federated recommender reconstruction. However, storing all historical parameter updates on resource-constrained personal devices is challenging and even infeasible. In light of this challenge, we propose a small-sized negative sampling method to reduce the number of item embedding updates and an importance-based update selection mechanism to store only important model updates. To evaluate the effectiveness of FRU, we propose an attack method to disturb FedRecs via a group of compromised users and use FRU to recover recommenders by eliminating these users' influence. Finally, we conduct experiments on two real-world recommendation datasets with two widely used FedRecs to show the efficiency and effectiveness of our proposed approaches.