PriDA:隱私維護去中心化匿名安全聚合於聯邦學習

PriDA: Privacy Preserving Decentralized Anonymous Secure Aggregation in Federated Learning

曾浩瑋(Hao-Wei Tseng)

指導教授 ： 廖婉君

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202102936

摘要

PriDA為確保隱私的安全聚合系統，於聯邦學習中計算聚合模型時同時保護每一位參與者之資料免於洩漏。只要參與者中存在一位非惡意攻擊者，任何參與者之資料皆能在安全聚合的過程中受保護。本文利用去中心化匿名以及資料混淆使得惡意攻擊者僅能獲得聚合過後之模型而不是特定參與者之私密資料。擴展了原始安全聚合之限制，本文透過動態聚合者選擇避免單點攻擊，並且透過去中心化匿名以及資料混淆放鬆了原先安全聚合需要多數參與者皆為誠實之參與者之限制，而是於不需要第三方信任機構存在之下，僅需要一位參與者為誠實之參與者即能保證每一位誠實參與者之隱私。

關鍵字

聯邦學習 ； 安全聚合 ； 去中心化匿名

並列摘要

We present a privacy-preserving secure aggregation system to compute global model in federated learning while preserving each participant’s sensitive data in the training process. As long as there exists one and another honest participant, privacy of honest participant’s sensitive data can be guaranteed in secure aggregation. We utilize decentralized anonymous and data obfuscation to make malicious attackers with corrupted participants only learn the aggregated model update instead of sensitive data of particular participant. Extending the primitive secure aggregation, we relax the privacy-preserving limitation from that the secure aggregation preserves privacy with the majority of honest participants to one and another honest participant by decentralized anonymity and prevent single point attack of aggregator by dynamic aggregators selection without trusted third party.

並列關鍵字

federated learning ； secure aggregation ； decentralized anonymous

參考文獻

[1] McMahan, Brendan, et al. "Communication-efficient learning of deep networks from decentralized data." Artificial Intelligence and Statistics. PMLR, 2017.

[2] Zhu, Ligeng, and Song Han. "Deep leakage from gradients." Federated learning.Springer, Cham, 2020. 17-31.

[3] Melis, Luca, et al. "Exploiting unintended feature leakage in collaborative learning." 2019 IEEE Symposium on Security and Privacy (SP). IEEE, 2019.

[4] Dwork, Cynthia, et al. "Calibrating noise to sensitivity in private data analysis." Theory of cryptography conference. Springer, Berlin, Heidelberg, 2006.

[5] Zhao, Yang, et al. "Local differential privacy-based federated learning for internet of things." IEEE Internet of Things Journal. 2020.