BatchCrypt: Efficient Homomorphic Encryption for Cross-Silo Federated Learning

By Chengliang Zhang, Suyi Li, Junzhe Xia, Wei Wang, Feng Yan, Yang Liu. USENIX Annual Technical Conference, 2020

927 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/c783cdc03a32e5094affa7eef710459aac599aaf

How researchers see this paper

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

1. Yang Liu
2. Qiang Yang
3. Xuefei Yin
4. Yanming Zhu
5. Tianjian Chen
6. Qian Xu
7. Jiankun Hu
8. Haokun Fang

Citations in the corpus

• Federated Learning: Opportunities and Challenges (2021, unknown): “Works such as [40] use cross-silo FL to develop their model.”
• FATE: An Industrial Grade Platform for Collaborative Learning With Data Protection (2021, methodology): “In (Zhang et al., 2020), a novel batch encryption algorithm is developed and improve the efficiency by 100× times.”
• Federated learning for predicting clinical outcomes in patients with COVID-19 (2021, background): “Encryption algorithms, such as homomorphic encryption 60 , maintain performance but may substantially increase message size and training time.”
• Federated Learning for Intrusion Detection System: Concepts, Challenges and Future Directions (2021, background): “Although there might be certain information loss, the bandwidth and the speed of data transfer are increased significantly improving the overall throughput and the performance of the deep learning model [Zhang et al.(2020)Zhang, Li, Xia, Wang, Yan, and Liu], [Li and Han(2019)].”
• Privacy Preserving Machine Learning with Homomorphic Encryption and Federated Learning (2021, methodology): “In 2020, Zhang et al. proposed a batchcrypt algorithm, which is based on the optimization of the FATE framework [46].”
• 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, ”
• FLAME: Taming Backdoors in Federated Learning (2022, background): “DP is a statistical approach that can be efficiently implemented, but it can only offer high privacy protection at the cost of a significant loss in accuracy due to the noise added to the models [6,61].”