Federated model with contrastive learning and adaptive control variates for human activity recognition
Regular Papers|Updated:2025-07-02
|
Federated model with contrastive learning and adaptive control variates for human activity recognition
Enhanced Publication
用于人体活动识别的基于对比学习与自适应变量控制的联邦模型
“In the field of human activity recognition, researchers have introduced FedCoad, a novel federated learning model that leverages contrastive learning with adaptive control variates to address data skewness among clients, enhancing model convergence and outperforming state-of-the-art FL algorithms on HAR benchmark datasets.”
Frontiers of Information Technology & Electronic EngineeringVol. 26, Issue 6, Pages: 896-911(2025)
Affiliations:
Department of Industrial and Management Engineering, Hankuk University of Foreign Studies, Yongin 17035, Republic of Korea
Ignatius IWAN, Bernardo Nugroho YAHYA, Seok-Lyong LEE. Federated model with contrastive learning and adaptive control variates for human activity recognition[J]. Frontiers of information technology & electronic engineering, 2025, 26(6): 896-911.
DOI:
Ignatius IWAN, Bernardo Nugroho YAHYA, Seok-Lyong LEE. Federated model with contrastive learning and adaptive control variates for human activity recognition[J]. Frontiers of information technology & electronic engineering, 2025, 26(6): 896-911. DOI: 10.1631/FITEE.2400797.
Federated model with contrastive learning and adaptive control variates for human activity recognitionEnhanced Publication