Improving Federated Learning With Quality-Aware User Incentive and Auto-Weighted Model Aggregation

• Published in: IEEE transactions on parallel and distributed systems Vol. 33; no. 12; pp. 4515 - 4529
• Main Authors: Deng, Yongheng, Lyu, Feng, Ren, Ju, Chen, Yi-Chao, Yang, Peng, Zhou, Yuezhi, Zhang, Yaoxue
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Agglomeration; Algorithms; Cognitive tasks; Collaborative work; Computational modeling; Data models; Edge computing; Electronic devices; Federated learning; incentive mechanism; learning quality; mobile computing; model aggregation; Nodes; Privacy; Resource management; Skewed distributions; Task analysis; Training; Training data

Federated learning enables distributed model training over various computing nodes, e.g., mobile devices, where instead of sharing raw user data, computing nodes can solely commit model updates without compromising data privacy. The quality of federated learning relies on the model updates contribut...