LDP-FL: Practical Private Aggregation in Federated Learning with Local Differential Privacy

By Lichao Sun, Jianwei Qian, Xun Chen, Philip S. Yu. 2020

278 citations recorded by the source corpus.

Training deep learning models on sensitive user data has raised increasing privacy concerns in many areas. Federated learning is a popular approach for privacy protection that collects the local gradient information instead of raw data. One way to achieve a strict privacy guarantee is to apply local differential privacy into federated learning. However, previous works do not give a practical solution due to two issues. First, the range difference of weights in different deep learning model layers has not been explicitly considered when applying local differential privacy mechanism. Second, the privacy budget explodes due to the high dimensionality of weights in deep learning models and many query iterations of federated learning. In this paper, we proposed a novel design of local differential privacy mechanism for federated learning to address the abovementioned issues. It makes the local weights update differentially private by adapting to the varying ranges at different layers of a deep neural network, which introduces a smaller variance of the estimated model weights, especially for deeper models. Moreover, the proposed mechanism bypasses the curse of dimensionality by parameter shuffling aggregation. A series of empirical evaluations on three commonly used datasets in prior differential privacy works, MNIST, Fashion-MNIST and CIFAR-10, demonstrate that our solution can not only achieve superior deep learning performance but also provide a strong privacy guarantee at the same time.

Subfields: Artificial Intelligence, Sociology and Political Science

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Lichao Sun
2. Lingjuan Lyu
3. Philip S. Yu
4. Qiang Yang
5. Han Yu
6. Jiankun Hu
7. Yanming Zhu
8. Xuefei Yin

In Sociology and Political Science

1. Lingjuan Lyu
2. Lichao Sun
3. Nguyen Binh Truong
4. Jun Zhao
5. Philip S. Yu
6. Kai Sun
7. Qiang Yang
8. Jiankun Hu

Citations in the corpus

• Federated Model Distillation with Noise-Free Differential Privacy (2020, methodology): “to differential privacy in federated learning (Geyer, Klein, and Nabi 2017; Bhowmick et al. 2018; Sun et al. 2020), none of them can achieve comparable performance on CIFAR with a small like us.”
• Privacy and Robustness in Federated Learning: Attacks and Defenses (2020, background): “porary work called LDP-FL [86] achieves better performance on both effectiveness and efficiency than [87] with the special communication design for deep learning approaches.”
• A Comprehensive Survey of Privacy-preserving Federated Learning (2021, methodology): “WHO RA1 [61, 196, 228] [26, 137, 218] [157, 169, 198] [144, 164, 199] [63, 70, 193] [16, 174, 191] [78, 110, 228] [52, 81, 216] [32, 56, 222] [30, 127, 173] RA2 [114, 175, 187] [215, 221, 224] [29, 111, 113] [13, 199] [3, 10, 40] [109, 155, 168] [45, 172, 193] [20, 52, 57] [56, 165, 197] [173, 199, ”
• Secure Deep Graph Generation with Link Differential Privacy (2020, methodology): “DP learning has also been widely adapted to generative models [Frigerio et al., 2019; Papernot et al., 2018; Sun et al., 2020b].”
• Privacy preservation in federated learning: An insightful survey from the GDPR perspective (2020, background): “User-level DP [95, 57, 48, 82, 10, 115] Membership Tracing”