vFedSec: Efficient Secure Aggregation for Vertical Federated Learning via Secure Layer
Keywords: Federated Learning, Secure Aggregation
TL;DR: We present vFedSec, a secure and efficient design for vertical FL. It implements an novel Secure Layer using SOTA security modules for secure aggregation, allowing intermediate outputs and gradients to be transmitted without compromising privacy.
Abstract: Most work in privacy-preserving federated learning (FL) has been focusing on horizontally partitioned datasets where clients share the same sets of features and can train complete models independently. However, in many interesting problems, individual data points are scattered across different clients/organizations in a vertical setting. Solutions for this type of FL require the exchange of intermediate outputs and gradients between participants, posing a potential risk of privacy leakage when privacy and security concerns are not considered. In this work, we present *vFedSec* - a novel design with an innovative *Secure Layer* for training vertical FL securely and efficiently using state-of-the-art security modules in secure aggregation. We theoretically demonstrate that our method does not impact the training performance while protecting private data effectively. Empirical results from extensive experiments substantiate this design producing secure training with negligible computation and communication overhead. Compared to widely-adopted homomorphic encryption (HE) methods, our method can obtain $\geq 690\times$ speedup and reduce communication costs by $\geq 9.6\times$.
Primary Area: general machine learning (i.e., none of the above)
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 2062
Loading