SCOPE-FL: Client Selection by Entropy Order

• Isaque O. Silva UFPA
• Carlos Vitelli UFPA
• Iago Medeiros UFPA

DOI: https://doi.org/10.5753/erin.2025.16155

Abstract

The increasing use of connected devices requires new methods to handle the quantity and privacy of shared data. Federated Learning (FL) emerges as a solution, enabling model training without directly sharing data, preserving the clients’ privacy. However, not all clients are equally useful for improving global models, making efficient client selection necessary. SCOPE-FL proposes a dynamic client selection mechanism, assigning weights to data entropy and dataset size to ensure a more efficient contribution to the global model. This is done by calculating a relevance score for each client based on these factors and adjusting the weights assigned to each client. SCOPE-FL uses the FedAvg method to aggregate local models, prioritizing clients with more relevant data. Tested with MNIST, SCOPE-FL outperformed traditional methods, showing an accuracy rate of over 60% after 12 rounds, reaching up to 80% after 22 rounds.

Keywords: Federated Learning, MNIST, Entropy

References

AbdulRahman, S., Tout, H., Ould-Slimane, H., Mourad, A., Talhi, C., and Guizani, M. (2020). A survey on federated learning: The journey from centralized to distributed on-site learning and beyond. IEEE Internet of Things Journal, 8(7):5476–5497.

de Souza, A. M., Bittencourt, L. F., Cerqueira, E., Loureiro, A. A., and Villas, L. A. (2023). Dispositivos, eu escolho vocês: Seleçao de clientes adaptativa para comunicaçao eficiente em aprendizado federado. In Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos (SBRC), pages 1–14. SBC.

Deng, Y., Lyu, F., Ren, J., Wu, H., Zhou, Y., Zhang, Y., and Shen, X. (2021). Auction: Automated and quality-aware client selection framework for efficient federated learning. IEEE Transactions on Parallel and Distributed Systems, 33(8):1996–2009.

McMahan, B., Moore, E., Ramage, D., Hampson, S., and y Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. In Artificial intelligence and statistics, pages 1273–1282. PMLR.

Orlandi, F. C., Dos Anjos, J. C., Santana, J. F. d. P., Leithardt, V. R., and Geyer, C. F. (2023). Entropy to mitigate non-iid data problem on federated learning for the edge intelligence environment. IEEE Access.

Pires, I. M., Marques, G., Garcia, N. M., Flórez-Revuelta, F., Canavarro Teixeira, M., Zdravevski, E., Spinsante, S., and Coimbra, M. (2020). Pattern recognition techniques for the identification of activities of daily living using a mobile device accelerometer. Electronics, 9(3):509.

Zhang, J., Hua, Y., Wang, H., Song, T., Xue, Z., Ma, R., and Guan, H. (2023). Fedala: Adaptive local aggregation for personalized federated learning. In AAAI Conference on Artificial Intelligence, volume 37, pages 11237–11244.