Analisis Trade-off Komunikasi-Komputasi pada Federated Averaging untuk Klasifikasi Penyakit Jantung
Abstract
Heart disease classification using Federated Learning (FL) offers a solution to enhance privacy in the medical field. Unfortunately, FL implementation is constrained by high communication costs that must be optimized while considering a low computational load. This research proposes an FL architecture combining Federated Averaging (FedAvg) with a Multi-Layer Perceptron (MLP) model for heart disease classification using the UCI Heart Disease dataset. This architecture is trained until the accuracy approaches the centralized training scenario as a threshold point across varying local epochs (E). Two metrics are considered in this study: communication cost and computational load, which are modeled using linear regression. Furthermore, the optimal point between both metrics is determined via the Weighted Sum Model (WSM) method. Test results show that communication cost decreases as the local epoch (E) increases, but is accompanied by an increase in computational load. Correlation trend modeling using linear regression yields an R² of 0.57 for communication cost with a negative slope and 0.76 for computational load with a positive slope, indicating a trade-off between both metrics. Finally, the optimal point between communication cost and computational load lies at E = 20, which has a minimum final WSM score of 0.2475. Keywords: federated learning, communication efficiency, FedAvg, IID, heart disease
Downloads
References
[1] C. W. Tsao et al., “Heart Disease and Stroke Statistics - 2023 Update: A Report from the American Heart Association,” 2023. doi: 10.1161/CIR.0000000000001123.
[2] A. A. Maulani, S. Winarno, J. Zeniarja, R. T. E. Putri, and A. N. Cahyani, “Comparison of Hyperparameter Optimization Techniques in Hybrid CNN-LSTM Model for Heart Disease Classification,” Sinkron, vol. 9, no. 1, 2024, doi: 10.33395/sinkron.v9i1.13219.
[3] A. G. Renauldy, “PERLINDUNGAN PRIVASI PADA PENGAMBILAN INFORMASI UNTUK BASIS DATA REKAM MEDIS,” Bandung, 2023.
[4] A. Horst et al., “Federated learning: a privacy-preserving approach to data-centric regulatory cooperation,” Frontiers in Drug Safety and Regulation, vol. 5, 2025, doi: 10.3389/fdsfr.2025.1579922.
[5] S. W. Herlambang, “FEDERATED LEARNING APPROACHES FOR IOT INTRUSION DETECTION BASED ON FEDAVG AND FEDPROX ON IID AND NON-IID DATA,” Bandung, 2025.
[6] H. Brendan McMahan, E. Moore, D. Ramage, S. Hampson, and B. Agüera y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017, 2017.
[7] Y. Luo, X. Jin, C. Xia, C. Xu, and Y. Sun, “Communication-Computation Co-Optimized Federated Learning for Efficient Large-Model Embedding Training,” Mathematics, vol. 13, no. 23, 2025, doi: 10.3390/math13233871.
[8] A. Imteaj, U. Thakker, S. Wang, J. Li, and M. H. Amini, “A Survey on Federated Learning for Resource-Constrained IoT Devices,” IEEE Internet Things J., vol. 9, no. 1, 2022, doi: 10.1109/JIOT.2021.3095077.
[9] K. Narmadha and P. Varalakshmi, “FedEff: efficient federated learning with optimal local epochs for heterogeneous clients,” Sci. Rep., 22672-1.
[10] J. Premsmith and H. Ketmaneechairat, “A predictive model for heart disease detection using data mining techniques,” Journal of Advances in Information Technology, vol. 12, no. 1, 2021, doi: 10.12720/jait.12.1.14-20.
[11] K. L. Du, C. S. Leung, W. H. Mow, and M. N. S. Swamy, “Perceptron: Learning, Generalization, Model Selection, Fault Tolerance, and Role in the Deep Learning Era,” 2022. doi: 10.3390/math10244730.
[12] A. Janosi, W. Steinbrunn, M. Pfisterer, and R. Detrano, “UCI Machine Learning Repository: Heart Disease Data Set,” Uci, 1988.
[13] C. M. Bishop and H. Bishop, Deep Learning: Foundations and Concepts. 2023. doi: 10.1007/978-3-031-45468-4.
[14] E. Saputra, E. R. Susanto, and K. Kunci, “Implementation of Deep Learning with Multilayer Perceptron (MLP) for Heart Disease Prediction Using the SMOTE-ENN Technique,” 2025. [Online]. Available: http://jurnal.polibatam.ac.id/index.php/JAIC
[15] A. Zhang, Z. C. Lipton, M. Li, and A. J. Smola, Dive into Deep Learning. Cambridge University Press, 2023.
[16] M. Nardi, L. Valerio, and A. Passarella, “Communication Costs Analysis of Unsupervised Federated Learning: an Anomaly Detection Scenario,” in CEUR Workshop Proceedings, 2023.
[17] G. James, D. Witten, T. Hastie, and R. Tibshirani, “An Introduction to Statistical Learning with Applications in R Second Edition,” Springer Texts, 2023.
[18] R. T. Marler and J. S. Arora, “The weighted sum method for multi-objective optimization: New insights,” Structural and Multidisciplinary Optimization, vol. 41, no. 6, 2010, doi: 10.1007/s00158-009-0460-7.
[19] P. Kairouz et al., “Advances and open problems in federated learning,” 2021. doi: 10.1561/2200000083.
[20] R. T. Marler and J. S. Arora, “Survey of multi-objective optimization methods for engineering Maximum objective function values,” Structural and Multidisciplinary Optimization, vol. 26, no. 6, 2004.
