Meta-Learning and Targeted Differential Privacy to Improve the Accuracy-Privacy Trade-off in Recommendations

TL;DR

This paper proposes a combined approach of targeted differential privacy and meta-learning to enhance recommendation accuracy while maintaining user privacy, outperforming standard methods.

Contribution

It introduces targeted DP at the data level and meta-learning at the model level to better balance privacy and accuracy in recommender systems.

Findings

Targeted DP reduces unnecessary data perturbation.

Meta-learning improves robustness to DP noise.

The combined approach outperforms standard privacy-preserving methods.

Abstract

Balancing differential privacy (DP) with recommendation accuracy is a key challenge in privacy-preserving recommender systems, since DP-noise degrades accuracy. We address this trade-off at both the data and model levels. At the data level, we apply DP only to the most stereotypical user data likely to reveal sensitive attributes, such as gender or age, to reduce unnecessary perturbation; we refer to this as targeted DP. At the model level, we use meta-learning to improve robustness to remaining DP-noise. This achieves a better trade-off between accuracy and privacy than standard approaches: Meta-learning improves accuracy and targeted DP leads to lower empirical privacy risk compared to uniformly applied DP and full DP baselines. Overall, our findings show that selectively applying DP at the data level together with meta-learning at the model level can effectively balance recommendation accuracy and user privacy.