Accepted author manuscript, 2.19 MB, PDF document
Available under license: CC BY: Creative Commons Attribution 4.0 International License
Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - Privacy-preserving Decentralized Federated Learning over Time-varying Communication Graph
AU - Lu, Yang
AU - Yu, Zhengxin
AU - Suri, Neeraj
PY - 2023/8/31
Y1 - 2023/8/31
N2 - Establishing how a set of learners can provide privacy-preserving federated learning in a fully decentralized (peer-to-peer, no coordinator) manner is an open problem. We propose the first privacy-preserving consensus-based algorithm for the distributed learners to achieve decentralized global model aggregation in an environment of high mobility, where participating learners and the communication graph between them may vary during the learning process. In particular, whenever the communication graph changes, the Metropolis-Hastings method [69] is applied to update the weighted adjacency matrix based on the current communication topology. In addition, the Shamir’s secret sharing scheme [61] is integrated to facilitate privacy in reaching consensus of the global model. The paper establishes the correctness and privacy properties of the proposed algorithm. The computational efficiency is evaluated by a simulation built on a federated learning framework with a real-world dataset.
AB - Establishing how a set of learners can provide privacy-preserving federated learning in a fully decentralized (peer-to-peer, no coordinator) manner is an open problem. We propose the first privacy-preserving consensus-based algorithm for the distributed learners to achieve decentralized global model aggregation in an environment of high mobility, where participating learners and the communication graph between them may vary during the learning process. In particular, whenever the communication graph changes, the Metropolis-Hastings method [69] is applied to update the weighted adjacency matrix based on the current communication topology. In addition, the Shamir’s secret sharing scheme [61] is integrated to facilitate privacy in reaching consensus of the global model. The paper establishes the correctness and privacy properties of the proposed algorithm. The computational efficiency is evaluated by a simulation built on a federated learning framework with a real-world dataset.
KW - Privacy
KW - Mobility
KW - Federated learning
KW - Decentralized aggregation
U2 - 10.1145/3591354
DO - 10.1145/3591354
M3 - Journal article
VL - 26
SP - 1
EP - 39
JO - ACM Transactions on Privacy and Security
JF - ACM Transactions on Privacy and Security
SN - 2471-2574
IS - 3
M1 - 33
ER -