Publicação
Federated learning with mobile nodes in V2X environments
| Resumo: | Federated Learning (FL) is a promising method for parameter normalization in machine learning (ML) models, especially when data privacy is crucial. However, there are significant constraints in FL solutions, particularly concerning the handling of the mobility of participating nodes in the parameter aggregation processes, with a substantial impact on Vehicular Ad-Hoc Networks (VANETs) within the scope of smart cities. To address this challenge, we present the Mobile Federated Learning System (MFLS), a lightweight microservices-based framework capable of operating on various types of devices (mobile and non-mobile). MFLS features an interface to integrate ML models to cooperate in the federated process without intrusion between the parties. MFLS manages the entire aggregation process, from instantiating services on mobile nodes to the final parameter updates in the involved ML models and the release of resources used in all participating nodes. Additionally, MFLS handles node mobility and ensures the proper execution of federated processes, even with nodes exiting at any stage of aggregation. To demonstrate the capabilities of MFLS, we used data collected through the communication infrastructure of Aveiro Tech City Living Lab (ATCLL), specifically the position of vehicles during their movement through the city. In our tests, we evaluated all phases of the aggregation process for mobile nodes, including creating the distributed services infrastructure, the interactive procedure for aggregation, and finally, after signaling parameter updates to ML models, executing the elimination of the temporary infrastructure created for the federated process. Our results show that even with intermittent connectivity in the Vehicular Ad- Hoc Network (VANET) of ATCLL, the MFLS system manages node mobility and effectively handles node availability during the aggregation of ML model parameters. |
|---|---|
| Autores principais: | Barreto, Bernardo de Francesco |
| Assunto: | Mobile federated learning FL on edge devices FL platforms Smart City |
| Ano: | 2023 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Aveiro |
| Idioma: | inglês |
| Origem: | RIA - Repositório Institucional da Universidade de Aveiro |
| Resumo: | Federated Learning (FL) is a promising method for parameter normalization in machine learning (ML) models, especially when data privacy is crucial. However, there are significant constraints in FL solutions, particularly concerning the handling of the mobility of participating nodes in the parameter aggregation processes, with a substantial impact on Vehicular Ad-Hoc Networks (VANETs) within the scope of smart cities. To address this challenge, we present the Mobile Federated Learning System (MFLS), a lightweight microservices-based framework capable of operating on various types of devices (mobile and non-mobile). MFLS features an interface to integrate ML models to cooperate in the federated process without intrusion between the parties. MFLS manages the entire aggregation process, from instantiating services on mobile nodes to the final parameter updates in the involved ML models and the release of resources used in all participating nodes. Additionally, MFLS handles node mobility and ensures the proper execution of federated processes, even with nodes exiting at any stage of aggregation. To demonstrate the capabilities of MFLS, we used data collected through the communication infrastructure of Aveiro Tech City Living Lab (ATCLL), specifically the position of vehicles during their movement through the city. In our tests, we evaluated all phases of the aggregation process for mobile nodes, including creating the distributed services infrastructure, the interactive procedure for aggregation, and finally, after signaling parameter updates to ML models, executing the elimination of the temporary infrastructure created for the federated process. Our results show that even with intermittent connectivity in the Vehicular Ad- Hoc Network (VANET) of ATCLL, the MFLS system manages node mobility and effectively handles node availability during the aggregation of ML model parameters. |
|---|