[en] Nowadays we are witnessing rapid development of the Internet of Things (IoT), machine learning, and cellular network technologies. They are key components to promote wireless networks beyond 5G (B5G). The plenty of data generated from numerous IoT devices, such as smart sensors and mobile devices, can be utilised to train intelligent models. But it still remains a challenge to efficiently utilise IoT networks and edge in B5G to conduct model training. In this paper, we propose a parallel training method which uses operators as scheduling units during training task assignment. Besides, we discuss a pebble-game-based memory-efficient optimisation in training. Experiments based on various real world network architectures show the flexibility of our proposed method and good performance compared with state of the art.
Research center :
CREMMI - Modélisation mathématique et informatique
Disciplines :
Computer science
Author, co-author :
Zhao, Jianxin ; School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China
VANDENHOVE, Pierre ; Université de Mons - UMONS > Faculté des Sciences > Service de Mathématiques effectives ; Université Paris-Saclay > Laboratoire Méthodes Formelles
Xu, Peng; School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China
Tao, Hao; China Ship Development and Design Center, Wuhan, China
Wang, Liang ; University of Cambridge, Cambridge, U.K.
Liu, Chi Harold ; School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China
Crowcroft, Jon ; University of Cambridge, Cambridge, U.K.
Language :
English
Title :
Parallel and Memory-Efficient Distributed Edge Learning in B5G IoT Networks
Publication date :
21 November 2022
Journal title :
IEEE Journal of Selected Topics in Signal Processing
ISSN :
1932-4553
eISSN :
1941-0484
Publisher :
Institute of Electrical and Electronics Engineers (IEEE)