Communication-Efficient and Utility-Enhanced Local Differential Privacy-Based Personalized Federated Compressed Learning | IEEE Journals & Magazine | IEEE Xplore

Communication-Efficient and Utility-Enhanced Local Differential Privacy-Based Personalized Federated Compressed Learning


Abstract:

With the deeper and broader research on federated learning (FL), several inescapable challenges arise when putting FL into practice. However, existing research works pred...Show More

Abstract:

With the deeper and broader research on federated learning (FL), several inescapable challenges arise when putting FL into practice. However, existing research works predominately concentrate on addressing one or two challenges. This paper seeks to provide a comprehensive exploration of four fundamental issues, namely privacy, utility, communication efficiency and data heterogeneity. To simultaneously address these issues, we propose a communication-efficient and utility-enhanced local differential privacy (LDP)-based personalized federated compressed learning (FCL) method, called CUEL-PFCL. First and foremost, a general FCL framework is proposed to compress local visual data (e.g., images) while preserving data learnability, which can provide a certain degree of visual-level privacy protection and improve the communication efficiency. Subsequently, an analytically tractable Gaussian differential privacy is applied to enhance the trade-off between privacy and utility. Meanwhile, compressed sensing and SIGNSGD are respectively used to compress and quantify model gradients to further reduce the communication overhead. Besides, we keep the head representation locally to reduce communication costs, achieve the privacy amplification effect and solve the issue of data heterogeneity. Theoretical privacy analysis, experimental simulations and comprehensive comparisons all demonstrate that CUEL-PFCL has four advantages, i.e., strong privacy, enhanced utility, efficient communication and various personalized models.
Published in: IEEE Transactions on Network Science and Engineering ( Volume: 12, Issue: 3, May-June 2025)
Page(s): 1776 - 1790
Date of Publication: 07 February 2025

ISSN Information:

Funding Agency:


I. Introduction

As The contradiction between the phenomenon of data silos and the requirement for data fusion becomes increasingly prominent, the advancement of Big Data-driven artificial intelligence applications is severely limited. Federated learning (FL), as a new distributed machine learning paradigm, ensures each participant's absolute control over its own sensitive data while allowing the collaborative training among participants [1], [2], [3]. Although FL avoids disclosing any local data, there is still a significant risk of privacy breaches. For example, attackers can utilize the captured communication gradients or parameters to reconstruct part of the sensitive data or even infer whether the mastered data originated from a particular participant [4], [5]. When implementing FL in realistic scenarios, some challenges cannot be ignored [6], [7], [8]. However, most existing researches focus on addressing one or two of these challenges. Therefore, here we aim to comprehensively explore four essential issues, i.e., data privacy, model utility, communication efficiency and data heterogeneity.

Contact IEEE to Subscribe

References

References is not available for this document.