Paper 2023/597
FedVS: Straggler-Resilient and Privacy-Preserving Vertical Federated Learning for Split Models
Abstract
In a vertical federated learning (VFL) system consisting of a central server and many distributed clients, the training data are vertically partitioned such that different features are privately stored on different clients. The problem of split VFL is to train a model split between the server and the clients. This paper aims to address two major challenges in split VFL: 1) performance degradation due to straggling clients during training; and 2) data and model privacy leakage from clients’ uploaded data embeddings. We propose FedVS to simultaneously address these two challenges. The key idea of FedVS is to design secret sharing schemes for the local data and models, such that information-theoretical privacy against colluding clients and curious server is guaranteed, and the aggregation of all clients’ embeddings is reconstructed losslessly, via decrypting computation shares from the non- straggling clients. Extensive experiments on various types of VFL datasets (including tabular, CV, and multi-view) demonstrate the universal advantages of FedVS in straggler mitigation and privacy protection over baseline protocols.
Metadata
- Available format(s)
-
PDF
- Category
- Cryptographic protocols
- Publication info
- Published elsewhere. ICML 2023
- Keywords
- Vertical Federated LearningStraggler MitigationPrivacy Protection
- Contact author(s)
-
songzeli8824 @ gmail com
dyao @ connect ust hk
jliu577 @ connect hkust-gz edu cn - History
- 2023-04-28: approved
- 2023-04-26: received
- See all versions
- Short URL
- https://ia.cr/2023/597
- License
-
CC BY-NC-SA
BibTeX
@misc{cryptoeprint:2023/597,
author = {Songze Li and Duanyi Yao and Jin Liu},
title = {{FedVS}: Straggler-Resilient and Privacy-Preserving Vertical Federated Learning for Split Models},
howpublished = {Cryptology {ePrint} Archive, Paper 2023/597},
year = {2023},
url = {https://eprint.iacr.org/2023/597}
}