Héber H. Arcolezi
LightNews LightNews
hharcolezi.bsky.social
Héber H. Arcolezi
@hharcolezi.bsky.social
15 followers 26 following 0 posts
Research Scientist at Inria Grenoble

Interested in #differentialprivacy, #MachineLearning, #MLprivacy, and #fairML. From 🇧🇷, now in 🇫🇷.
Posts Replies Media Videos
Reposted by Héber H. Arcolezi
dppapers.bsky.social
Revisiting Locally Differentially Private Protocols: Towards Better Trade-offs in Privacy, Utility, and Attack Resistance
Héber H. Arcolezi, Sébastien Gambs
http://arxiv.org/abs/2503.01482
Revisiting Locally Differentially Private Protocols: Towards Better Trade-offs in Privacy, Utility, and Attack Resistance Héber H. Arcolezi, Sébastien Gambs http://arxiv.org/abs/2503.01482 Local Differential Privacy (LDP) offers strong privacy protection, especially in settings in which the server collecting the data is untrusted. However, designing LDP mechanisms that achieve an optimal trade-off between privacy, utility, and robustness to adversarial inference attacks remains challenging. In this work, we introduce a general multi-objective optimization framework for refining LDP protocols, enabling the joint optimization of privacy and utility under various adversarial settings. While our framework is flexible enough to accommodate multiple privacy and security attacks as well as utility metrics, in this paper we specifically optimize for Attacker Success Rate (ASR) under distinguishability attack as a measure of privacy and Mean Squared Error (MSE) as a measure of utility. We systematically revisit these trade-offs by analyzing eight state-of-the-art LDP protocols and proposing refined counterparts that leverage tailored optimization techniques. Experimental results demonstrate that our proposed adaptive mechanisms consistently outperform their non-adaptive counterparts, reducing ASR by up to five orders of magnitude while maintaining competitive utility. Analytical derivations also confirm the effectiveness of our mechanisms, moving them closer to the ASR-MSE Pareto frontier.
March 5, 2025 at 6:43 AM