PAC-DP: Personalized Adaptive Clipping for Differentially Private Federated Learning

Computer Science > Cryptography and Security [Submitted on 25 Mar 2026] PAC-DP: Personalized Adaptive Clipping for Differentially Private Federated Learning Hao Zhou, Siqi Cai, Hua Dai, Geng Yang, Jing Luo, Hui Cai Differential privacy (DP) is crucial for safeguarding sensitive client information in federated learning (FL), yet traditional DP-FL methods rely predominantly on fixed gradient clipping thresholds. Such static clipping neglects significant client heterogeneity and varying privacy sensitivities, which may lead to an unfavorable privacy-utility trade-off. In this paper, we propose PAC-DP, a Personalized Adaptive Clipping framework for federated learning under record-level local differential privacy. PAC-DP introduces a Simulation-CurveFitting approach leveraging a server-hosted public proxy dataset to learn an effective mapping between personalized privacy budgets epsilon and gradient clipping thresholds C, which is then deployed online with a lightweight round-wise schedule. This design enables budget-conditioned threshold selection while avoiding data-dependent tuning during training. We provide theoretical analyses establishing convergence guarantees under the per-example clipping and Gaussian perturbation mechanism and a reproducible privacy accounting procedure. Extensive evaluations on multiple FL benchmarks show that PAC-DP surpasses conventional fixed-threshold approaches under matched privacy budgets, improving accuracy by up to 26% and accelerating convergence by up to 45.5% in our evaluated settings. Comments: *Corresponding author: Hua Dai. 15 pages, 13 figures Subjects: Cryptography and Security (cs.CR) ACM classes: K.4.1; I.2.11; G.1.6 Cite as: arXiv:2603.24003 [cs.CR]   (or arXiv:2603.24003v1 [cs.CR] for this version)   https://doi.org/10.48550/arXiv.2603.24003 Focus to learn more Submission history From: Hao Zhou [view email] [v1] Wed, 25 Mar 2026 07:06:33 UTC (670 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CR < prev   |   next > new | recent | 2026-03 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)