Repository logo

Infoscience

  • English
  • French
Log In
Logo EPFL, École polytechnique fédérale de Lausanne

Infoscience

  • English
  • French
Log In
  1. Home
  2. Academic and Research Output
  3. Conferences, Workshops, Symposiums, and Seminars
  4. POSEIDON: Privacy-Preserving Federated Neural Network Learning
 
conference paper

POSEIDON: Privacy-Preserving Federated Neural Network Learning

Sav, Sinem  
•
Pyrgelis, Apostolos  
•
Troncoso-Pastoriza, Juan Ramon  
Show more
January 1, 2021
28Th Annual Network And Distributed System Security Symposium (Ndss 2021)
28th Annual Network and Distributed System Security Symposium (NDSS)

In this paper, we address the problem of privacy-preserving training and evaluation of neural networks in an N-party, federated learning setting. We propose a novel system, POSEIDON, the first of its kind in the regime of privacy-preserving neural network training. It employs multiparty lattice-based cryptography to preserve the confidentiality of the training data, the model, and the evaluation data, under a passive-adversary model and collusions between up to N - 1 parties. To efficiently execute the secure backpropagation algorithm for training neural networks, we provide a generic packing approach that enables Single Instruction, Multiple Data (SIMD) operations on encrypted data. We also introduce arbitrary linear transformations within the cryptographic bootstrapping operation, optimizing the costly cryptographic computations over the parties, and we define a constrained optimization problem for choosing the cryptographic parameters. Our experimental results show that POSEIDON achieves accuracy similar to centralized or decentralized non-private approaches and that its computation and communication overhead scales linearly with the number of parties. POSEIDON trains a 3-layer neural network on the MNIST dataset with 784 features and 60K samples distributed among 10 parties in less than 2 hours.

  • Details
  • Metrics
Type
conference paper
DOI
10.14722/ndss.2021.24119
Web of Science ID

WOS:000680821200077

Author(s)
Sav, Sinem  
Pyrgelis, Apostolos  
Troncoso-Pastoriza, Juan Ramon  
Froelicher, David  
Bossuat, Jean-Philippe  
Sousa, Joao Sa  
Hubaux, Jean-Pierre  
Date Issued

2021-01-01

Publisher

INTERNET SOC

Publisher place

Reston

Published in
28Th Annual Network And Distributed System Security Symposium (Ndss 2021)
ISBN of the book

1-891562-66-5

Subjects

Computer Science, Information Systems

•

Computer Science, Theory & Methods

•

Computer Science

Editorial or Peer reviewed

REVIEWED

Written at

EPFL

EPFL units
LDS  
Event nameEvent placeEvent date
28th Annual Network and Distributed System Security Symposium (NDSS)

ELECTR NETWORK

Feb 21-25, 2021

Available on Infoscience
September 11, 2021
Use this identifier to reference this record
https://infoscience.epfl.ch/handle/20.500.14299/181222
Logo EPFL, École polytechnique fédérale de Lausanne
  • Contact
  • infoscience@epfl.ch

  • Follow us on Facebook
  • Follow us on Instagram
  • Follow us on LinkedIn
  • Follow us on X
  • Follow us on Youtube
AccessibilityLegal noticePrivacy policyCookie settingsEnd User AgreementGet helpFeedback

Infoscience is a service managed and provided by the Library and IT Services of EPFL. © EPFL, tous droits réservés