Banca de DEFESA: JOHN LUCAS RODRIGUES PORTILHO DE SOUSA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
DISCENTE: JOHN LUCAS RODRIGUES PORTILHO DE SOUSA
DATA: 13/08/2024
HORA: 09:00
LOCAL: GERCOM
TÍTULO:

Entropy-based Client Selection Strategy for  Federated Learning over Vehicular Network Environments

PALAVRAS-CHAVES:

Federated Learning; Vehicular Networks; Connected and Autonomous Vehicles; Client Selection; Entropy; Client Failures

PÁGINAS: 53
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Ciência da Computação
SUBÁREA: Sistemas de Computação
RESUMO:

Federated Learning (FL) emerges as a promising solution to enable collaborative model training for autonomous vehicles while preserving privacy and addressing communication overhead issues. Efficient client selection for participation in the training process remains challenging, especially in scenarios with statistical heterogeneity of data distribution and client failure events. Client failure, an uncontrollable event during training, reduces accuracy, convergence, and speed. This master thesis introduces an entropy-based client selection mechanisms for FL over Vehicular Network environments with client failure and non-IID data distributions. The proposed method is compared to a random selection mechanism in both IID and non-IID scenarios, as well as scenarios with random client drops. The results demonstrate that entropy-based selection outperforms other methods regarding training loss, accuracy, and Area Under the Roc Curve, particularly in high client dropout and non-IID scenarios. These findings highlight the importance of considering entropy data for client selection to address the challenges posed by client failure and statistical heterogeneity in  FL over Vehicular Network.

MEMBROS DA BANCA:
Presidente - 1700540 - EDUARDO COELHO CERQUEIRA
Interno - 2244893 - DENIS LIMA DO ROSARIO
Interno - 3439716 - DIEGO LISBOA CARDOSO
Externo ao Programa - 3399038 - IAGO LINS DE MEDEIROS