Defesa de Dissertação de Mestrado – Flavio Vieira
Orientador: Carlos Alberto Vieira Campos
Título da dissertação/tese: reducing costs using normalization in federated learning in heterogeneous data distributions
Resumo: In an increasingly connected world, technologies such as smartphones, 5G, drones, the Internet of Things, and Smart Cities bring new challenges and opportunities. The increase in data collected by these devices and their ease of access allows the use of machine learning techniques to provide intelligent and quality services. Considering these services’ distributed access to data, using Federated Learning is a great option that allows for decentralized machine learning processing with greater security. However, client access to heterogeneously distributed data impacts federated learning, reducing test accuracy and increasing communication costs. To address these problems, we present a new method called Federated Learning with Weight Standardization on Convolutional Neural Networks (FedWS) that uses standardization on weights in local training on convolutional neural networks, optimizing the training gradients and reducing the impact of weight divergence caused by heterogeneous distributions on federated training tasks. Results showed that our method got superior results on image classification tasks in the order of 1.36% to 5.0% of test accuracy at different levels of heterogeneity. Their behavior showed a reduction in the effects of divergence at higher levels of heterogeneity and communication cost reduction ranging from 20% to 100%, being more suitable for use in mobile devices with computational resource limitations.
Palavras-chave: Federated Learning, Heterogeneous Data, Data Transformation, Image Classification.
Banca examinadora:
Carlos Alberto Vieira Campos
Carlos Eduardo Ribeiro de Mello – Unirio
Aline Marins Paes carvalho – IC/UFF
Data/hora de defesa: 01/04/2024 13:30
