EMDG-FL: Enhanced Malicious Model Detection based on Genetic Algorithm for Federated Learning

Federated learning (FL) enables collaborative machine learning among multiple devices without sharing private data. However, FL systems are vulnerable to poisoning attacks where malicious participants send malicious model updates to compromise the global model's accuracy. To enhance malicious m...

Full description

Saved in:
Bibliographic Details
Main Authors: Atia, Okba Ben, Samara, Mustafa Al, Bennis, Ismail, Gaber, Jaafar, Abouaissa, Abdelhafid, Lorenz, Pascal
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Accuracy Rate (ACC)
Attack Success Rate (ASR)
Collaboration
Federated learning
Federated Learning (FL)
Genetic Algorithm (GA)
Genetic algorithms
Loss Rate (LR)
Measurement
poisoning attacks
Simulation
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Federated learning (FL) enables collaborative machine learning among multiple devices without sharing private data. However, FL systems are vulnerable to poisoning attacks where malicious participants send malicious model updates to compromise the global model's accuracy. To enhance malicious model detection, we propose an EMDG-FL approach that optimizes the threshold used to identify attacks through a Genetic Algorithm (GA). The threshold indicates the degree of divergence between benign and malicious model updates. A tightly tuned threshold improves detection efficiency by reducing false positives and negatives. Our approach also includes a comparison study evaluating EMDG-FL against other defenses from literature across metrics like Accuracy Rate (ACC), Attack Success Rate (ASR) and Loss Rate (LR). Simulation results using two datasets demonstrate that EMDG-FL outperforms prior works in detecting poisoning attacks in FL. The optimized threshold calculation enables more precise and efficient identification of malicious models.
ISSN:1558-2612
DOI:10.1109/WCNC57260.2024.10570752