PAC-DP: Personalized Adaptive Clipping for Differentially Private Federated Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914420520648704 |
|---|---|
| author | Zhou, Hao Cai, Siqi Dai, Hua Yang, Geng Luo, Jing Cai, Hui |
| author_facet | Zhou, Hao Cai, Siqi Dai, Hua Yang, Geng Luo, Jing Cai, Hui |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24003 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PAC-DP: Personalized Adaptive Clipping for Differentially Private Federated Learning Zhou, Hao Cai, Siqi Dai, Hua Yang, Geng Luo, Jing Cai, Hui Cryptography and Security K.4.1; I.2.11; G.1.6 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. |
| title | PAC-DP: Personalized Adaptive Clipping for Differentially Private Federated Learning |
| topic | Cryptography and Security K.4.1; I.2.11; G.1.6 |
| url | https://arxiv.org/abs/2603.24003 |