Paper 2022/1544

Towards Efficient Decentralized Federated Learning

Christodoulos Pappas, Hong Kong University of Science and Technology
Dimitrios Papadopoulos, Hong Kong University of Science and Technology
Dimitris Chatzopoulos, University College Dublin
Eleni Panagou, University of Thessaly
Spyros Lalis, University of Thessaly
Manolis Vavalis, University of Thessaly
Abstract

We focus on the problem of efficiently deploying a federated learning training task in a decentralized setting with multiple aggregators. To that end, we introduce a number of improvements and modifications to the recently proposed IPLS protocol. In particular, we relax its assumption for direct communication across participants, using instead indirect communication over a decentralized storage system, effectively turning it into a partially asynchronous protocol. Moreover, we secure it against malicious aggregators (that drop or alter data) by relying on homomorphic cryptographic commitments for efficient verification of aggregation. We implement the modified IPLS protocol and report on its performance and potential bottlenecks. Finally, we identify important next steps for this line of research.

Metadata
Available format(s)
PDF
Category
Applications
Publication info
Preprint.
Keywords
Federated Learning Decentralized Storage InterPlanetary File System Verifiable Aggregation Homomorphic Commitments
Contact author(s)
cpappas @ connect ust hk
dipapado @ cse ust hk
dimitris chatzopoulos @ ucd ie
epanagou @ e-ce uth gr
lalis @ uth gr
mav @ uth gr
History
2022-11-08: approved
2022-11-07: received
See all versions
Short URL
https://ia.cr/2022/1544
License
No rights reserved
CC0

BibTeX

@misc{cryptoeprint:2022/1544,
      author = {Christodoulos Pappas and Dimitrios Papadopoulos and Dimitris Chatzopoulos and Eleni Panagou and Spyros Lalis and Manolis Vavalis},
      title = {Towards Efficient Decentralized Federated Learning},
      howpublished = {Cryptology ePrint Archive, Paper 2022/1544},
      year = {2022},
      note = {\url{https://eprint.iacr.org/2022/1544}},
      url = {https://eprint.iacr.org/2022/1544}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.