TY - GEN
T1 - Meta-Learning and Targeted Differential Privacy to Improve the Accuracy-Privacy Trade-off in Recommendations
AU - Müllner, Peter
AU - Kowald, Dominik
AU - Schedl, Markus
AU - Lex, Elisabeth
N1 - Publisher Copyright:
© 2026 Copyright held by the owner/author(s).
PY - 2026/6/7
Y1 - 2026/6/7
N2 - 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.
AB - 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.
KW - Accuracy-Privacy Trade-off
KW - Differential Privacy
KW - Meta-Learning
KW - Recommender Systems
KW - Stereotypical User Preferences
UR - https://www.scopus.com/pages/publications/105042896532
U2 - 10.1145/3774935.3812709
DO - 10.1145/3774935.3812709
M3 - Conference paper
AN - SCOPUS:105042896532
T3 - UMAP 2026 - Proceedings of the 34th ACM International Conference on User Modeling, Adaptation and Personalization
SP - 517
EP - 520
BT - UMAP 2026 - Proceedings of the 34th ACM International Conference on User Modeling, Adaptation and Personalization
PB - Association for Computing Machinery (ACM)
T2 - 34th ACM International Conference on User Modeling, Adaptation and Personalization, UMAP 2026
Y2 - 8 June 2026 through 11 June 2026
ER -