$\lambda$-SecAgg: Partial Vector Freezing for Lightweight Secure Aggregation in Federated Learning
Keywords: Secure aggregation, Federated learning
TL;DR: A plug-and-play module to compress computation overhead in secure aggregation protocols for federated learning, requiring only $\frac{1}{\lambda}$ of the original vector to participate in the protocol.
Abstract: Secure aggregation of user update vectors (e.g. gradients) has become a critical issue in the field of federated learning. Many Secure Aggregation Protocols (SAPs) face exorbitant computation costs, severely constraining their applicability. Given the observation that a considerable portion of SAP's computation burden stems from processing each entry in the private vectors, we propose Partial Vector Freezing (PVF), a portable module for compressing computation costs without introducing additional communication overhead. $\lambda$-SecAgg, which integrates SAP with PVF, "freezes" a substantial portion of the private vector through specific transformations, requiring only $\frac{1}{\lambda}$ of the original vector to participate in SAP. Eventually, users can "thaw" the public sum of the "frozen entries" by the result of SAP. To avoid potential privacy leakage, we devise Disrupting Variables Element for PVF. We demonstrate that PVF can seamlessly integrate with various SAPs and it poses no threat to user privacy in the semi-honest and active adversary settings. We include $7$ baselines, encompassing $5$ distinct types of masking schemes, and explore the acceleration effects of PVF on these SAPs. Empirical investigations indicate that when $\lambda=100$, PVF yields up to $99.5\times$ speedup and up to $32.3\times$ communication reduction.
Primary Area: alignment, fairness, safety, privacy, and societal considerations
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/2025/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: 4067
Loading