Federated Learning based on Multiple Decision Trees for Cooperative IoT Edge Computing

• Lucas Barbosa UFPA
• Rodrigo Ferreira UFPA
• Bruno L. Dalmazo FURG
• Glauco Gonçalves UFPA
• Adonis Leal New Mexico Institute of Mining and Technology
• André Riker UFPA

DOI: https://doi.org/10.5753/courb.2025.9496

Abstract

Internet of Things (IoT) have relied on edge computing nodes to decentralize computation and to bring more processing power near the IoT devices, such as sensors and actuators. IoT edge computing nodes have more data processing power and energy resources than regular IoT devices that aim to monitor and actuate on the environment. However, in general, IoT edge computing nodes are not designed for intensive Machine Learning (ML) training or to host large ML models. In the current IoT network architectures, there are multiple IoT edge computing nodes strategically located near a large number of IoT devices, where each of the IoT edge computing node has access to part of the data produced by the whole IoT network. In this scenario, each IoT edge computing node runs lightweight ML models in its local dataset. In this paper, we propose a solution, called FEDT (FEderated Decision Tree), that aggregates the learning produced by multiple decision trees from cooperative IoT edge nodes, following the federated learning principles. We present four different federated learning strategies and demonstrate that FEDT can achieve around 80% of a centralized ML model in terms of Pearson correlation.

Keywords: Federated Learning, IoT, Decision Tree

References

Berman, J. J. (2016). Chapter 4 - Understanding your data. In Berman, J. J. (Ed.), Data Simplification (pp. 135–187). Morgan Kaufmann.

Candanedo, L. M., Feldheim, V., & Deramaix, D. (2017). Data driven prediction models of energy use of appliances in a low-energy house. Energy and Buildings, 140, 81–97.

Grinsztajn, L., Oyallon, E., & Varoquaux, G. (2022). Why do tree-based models still outperform deep learning on typical tabular data? Advances in Neural Information Processing Systems, 35, 507–520.

Hauschild, A.-C., Lemanczyk, M., Matschinske, J., Frisch, T., Zolotareva, O., Holzinger, A., Baumbach, J., & Heider, D. (2022). Federated Random Forests can improve local performance of predictive models for various healthcare applications. Bioinformatics, 38(8), 2278–2286.

Huang, P., Li, D., & Yan, Z. (2023). Wireless federated learning with asynchronous and quantized updates. IEEE Communications Letters, 27(9), 2393–2397.

Ji, Y., & Chen, L. (2023). Fedqnn: A computation–communication-efficient federated learning framework for IoT with low-bitwidth neural network quantization. IEEE Internet of Things Journal, 10(3), 2494–2507.

Leite, L., Santo, Y., Dalmazo, B. L., & Riker, A. (2024). Federated learning under attack: Improving gradient inversion for batch of images. arXiv preprint arXiv:2409.17767.

Lindskog, W., & Prehofer, C. (2023). A federated learning benchmark on tabular data: Comparing tree-based models and neural networks. In 2023 Eighth International Conference on Fog and Mobile Edge Computing (FMEC) (pp. 239–246).

Markovic, T., Leon, M., Buffoni, D., & Punnekkat, S. (2022). Random forest based on federated learning for intrusion detection. In Maglogiannis, I., Iliadis, L., Macintyre, J., & Cortez, P. (Eds.), Artificial Intelligence Applications and Innovations (pp. 132–144). Cham: Springer International Publishing.

Maseer, Z. K., Yusof, R., Bahaman, N., Mostafa, S. A., & Foozy, C. F. M. (2021). Benchmarking of machine learning for anomaly based intrusion detection systems in the CICIDS2017 dataset. IEEE Access, 9, 22351–22370.

Nettleton, D. (2014). Chapter 6 - Selection of variables and factor derivation. In Nettleton, D. (Ed.), Commercial Data Mining (pp. 79–104). Morgan Kaufmann.

Reddi, S., Charles, Z., Zaheer, M., Garrett, Z., Rush, K., Konecny, J., Kumar, S., & McMahan, H. B. (2020). Adaptive federated optimization. arXiv preprint arXiv:2003.00295.

Riker, A., Mota, R., Rosário, D., Pereira, V., & Curado, M. (2022). Autonomic management of group communication for Internet of Things applications. International Journal of Communication Systems, 35(11), e5200.

Santo, Y., Abelém, A., Riker, A., & Tsiropoulou, E. E. (2025). Integrating data privacy and energy management in smart cities with partial sustainable IoT networks. IEEE Access.

Shen, T., Mishra, C. S., Sampson, J., Kandemir, M. T., & Narayanan, V. (2022). An efficient edge-cloud partitioning of random forests for distributed sensor networks. IEEE Embedded Systems Letters.

Silva, M., Riker, A., Torrado, J., Santos, J., & Curado, M. (2021). Extending energy neutral operation in Internet of Things. IEEE Internet of Things Journal, 9(10), 7510–7524.

Singhal, P., Pandey, S. R., & Popovski, P. (2024). Greedy Shapley client selection for communication-efficient federated learning. IEEE Networking Letters, 1–1.

Ullah, A., Anwar, S. M., Li, J., Nadeem, L., Mahmood, T., Rehman, A., & Saba, T. (2024). Smart cities: The role of Internet of Things and machine learning in realizing a data-centric smart environment. Complex & Intelligent Systems, 10(1), 1607–1637.

Yu, R., & Li, P. (2021). Toward resource-efficient federated learning in mobile edge computing. IEEE Network, 35(1), 148–155.