TY - GEN
T1 - Exploiting diversity via importance-aware user scheduling for fast edge learning
AU - Liu, Dongzhu
AU - Zhu, Guangxu
AU - Zhang, Jun
AU - Huang, Kaibin
N1 - Funding Information:
V. CONCLUDING REMARKS In this paper, we proposed a novel scheduling scheme, namely importance-aware scheduling, for wireless data acquisition in edge learning systems. The scheme intelligently makes a joint channel-and-data selection for training data uploading so as to accelerate learning speed. Comprehensive experiments using real datasets substantiate the performance gain by exploiting two-fold multi-user diversity, namely multiuser data-and-channel diversity. At a higher level, the work contributes the new principle of exploiting data importance to improve the efficiency of multiuser data acquisition for distributed edge learning. Besides raw data acquisition, another interesting direction is the acquisition of learning relevant information in a federated learning framework, e.g., gradient updates and model updates, and the design therein involves changing the data-importance metrics to gradient divergence and model variance. VI. ACKNOWLEDGEMENT The work was supported in part by Hong Kong Research Grants Council under the Grants 17208319 and 17209917. Dr. G. Zhu was supported by an internal fund of Shenzhen Research Institute of Big Data (Project ID J00120190020).
Publisher Copyright:
© 2020 IEEE.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/6
Y1 - 2020/6
N2 - With the prevalence of intelligent mobile applications, edge learning is emerging as a promising technology for powering fast intelligence acquisition for edge devices from distributed data generated at the network edge. One critical task of edge learning is to efficiently utilize the limited radio resource to acquire data samples for model training at an edge server. In this paper, we develop a novel user scheduling algorithm for data acquisition in edge learning, called (data) importance-aware scheduling. A key feature of this scheduling algorithm is that it takes into account the informativeness of data samples, besides communication reliability. Specifically, the scheduling decision is based on a data importance indicator (DII), elegantly incorporating two importance metrics from communication and learning perspectives, i.e., the signal-to-noise ratio (SNR) and data uncertainty. We derive an explicit expression for this indicator targeting the classic classifier of support vector machine (SVM), where the uncertainty of a data sample is measured by its distance to the decision boundary. As demonstrated via experiments using real datasets, the proposed importance-aware scheduling method can exploit the two-fold multi-user diversity, namely the diversity in both the multiuser channels and the distributed data samples. This leads to faster model convergence than the conventional scheduling schemes that exploit only a single type of diversity.
AB - With the prevalence of intelligent mobile applications, edge learning is emerging as a promising technology for powering fast intelligence acquisition for edge devices from distributed data generated at the network edge. One critical task of edge learning is to efficiently utilize the limited radio resource to acquire data samples for model training at an edge server. In this paper, we develop a novel user scheduling algorithm for data acquisition in edge learning, called (data) importance-aware scheduling. A key feature of this scheduling algorithm is that it takes into account the informativeness of data samples, besides communication reliability. Specifically, the scheduling decision is based on a data importance indicator (DII), elegantly incorporating two importance metrics from communication and learning perspectives, i.e., the signal-to-noise ratio (SNR) and data uncertainty. We derive an explicit expression for this indicator targeting the classic classifier of support vector machine (SVM), where the uncertainty of a data sample is measured by its distance to the decision boundary. As demonstrated via experiments using real datasets, the proposed importance-aware scheduling method can exploit the two-fold multi-user diversity, namely the diversity in both the multiuser channels and the distributed data samples. This leads to faster model convergence than the conventional scheduling schemes that exploit only a single type of diversity.
UR - http://www.scopus.com/inward/record.url?scp=85090277739&partnerID=8YFLogxK
U2 - 10.1109/ICCWorkshops49005.2020.9145034
DO - 10.1109/ICCWorkshops49005.2020.9145034
M3 - Conference article published in proceeding or book
AN - SCOPUS:85090277739
T3 - 2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020 - Proceedings
BT - 2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020
Y2 - 7 June 2020 through 11 June 2020
ER -