Byzantine-robust federated learning performance evaluation via distance-statistical aggregations

Francesco Colosimo; Giovanni Rocca

doi:10.1117/12.3017007

7 June 2024 Byzantine-robust federated learning performance evaluation via distance-statistical aggregations

Francesco Colosimo, Giovanni Rocca

Proceedings Volume 13054, Assurance and Security for AI-enabled Systems; 130540T (2024) https://doi.org/10.1117/12.3017007
Event: SPIE Defense + Commercial Sensing, 2024, National Harbor, Maryland, United States

Conference Poster

Abstract

Federated Learning (FL) has emerged as a prominent branch of Machine Learning due to the increasing prevalence of mobile computing and IoT technologies. Unlike centralized systems, in FL the devices often operate beyond the confines of centralized protection mechanisms. Consequently, the adoption of this methodology gives rise to various security concerns, including data leakage, communication vulnerabilities, and poisoning. In this paper we propose new distance-statistical aggregation algorithms that provide robustness against Byzantine failures, and we compare them with the well-known FedAvg on a set of simulations that recreate realistic scenarios. Achieved results demonstrate the functionality of the solutions in terms of efficiency and accuracy.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Francesco Colosimo and Giovanni Rocca "Byzantine-robust federated learning performance evaluation via distance-statistical aggregations", Proc. SPIE 13054, Assurance and Security for AI-enabled Systems, 130540T (7 June 2024); https://doi.org/10.1117/12.3017007

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available