Protecting federated learning from malicious attacks using consensus technique

doi:10.1016/j.asoc.2025.114299

Protecting federated learning from malicious attacks using consensus technique

2025-11-20

0

PRE

AI

Frederica Free-Nelson

DOI：10.1016/j.asoc.2025.114299

Original

Original request for help

Save

Abstract

En 中文

• Vulnerability assessment of federated learning (FL) in malicious attack scenarios. • Resilient global model aggregation by a consensus confirmation technique. • Quantitative analysis of security and overhead performance for FL algorithms with consensus confirmation. • Experimental demonstration of the FL algorithms under malicious attacks.

Protecting federated learning from malicious attacks using consensus technique

Abstract

Journal

Organization