Call for Papers

Recently, with the maturity of edge-cloud computing and the large amount of data generated in the edge, we have witnessed an increasing number of applications conducting collaborative learning and data analytics in networked edge systems. The market share of collaborative data analytics in the edge is USD 20 Billion and the projection in 2030 will be more than USD 60 Billion. On the other end of the spectrum, there are also fast-growing concerns on privacy on using the data in the edge, which belong to diverse owners. In 2020, the fine topped USD1 Billion for data privacy violations and if there are no proper solutions, the fine can reach USD20 Billion in 2030.

Federated learning (FL) and federated analytics (FA), coined together as federated optimizations by Google, are new distributed computing techniques to address such a mismatch. In FL and FA, raw data are kept local and only the focused updates (weights or data insights) generated from local analytics are sent to a cloud server for result aggregation. FL focuses more on deep neural network model training for predictive tasks; FA expands this federated paradigm to all data analytics operations.

The advancement of federated optimizations has created promising opportunities to enable increasingly complex networked systems, involving diverse users, service providers, network operators, to become more intelligent and autonomous in network management with privacy protection. Despite the potential and opportunities that federated optimizations bring about to facilitate intelligent network management in privacy-demanding environments, there remain many challenges. FL/FA services naturally rely on networking. The limited bandwidth resource and heterogeneous system constraints put additional constraints on the design space for federated optimization algorithms. On the other hand, designing privacy-preserving federated monitoring, modelling, and overall control of ever-increasing complex networks composed of a heterogeneous set of entities remain unrealized.

This Special Issue (SI) aims to look into the fundamental challenges and opportunities that intersect between federated optimizations and networked systems. We welcome submissions from both academia and industry that address issues from both the perspective on network and system architecture and supports for FL/FA services as well as the perspective on federated optimizations to improve networked systems. Topics include, but are not limited to:

• Federated optimization framework for network data
• Network and system architecture for federated optimizations
• Communication and networking technologies for federated optimizations
• Resource-efficient FL/FA optimizations in heterogeneous networked systems
• Federated telemetry, traffic classification, clustering analytics for networked systems
• Federated responsible AI and federated explainable AI
• Federated anomaly detection, intrusion detection, vulnerability prevention for network security
• Federated channel estimation, user behavior analysis, mobility characterization, QoE analysis in wireless networks
• FL/FA-aided multi-agent reinforcement learning, bandits for network scheduling and control
• FL/FA for emerging networked applications and systems such as 6G, autonomous vehicle systems, XR, digital twin, smart health
• Efficient local differential privacy, secure multiparty computation, trusted execution environment for FL/FA in networked systems
• Incentive mechanism design, pricing, business models for FL/FA services in networked systems
• Experimental testbeds, case studies, performance evaluations of FL/FA in networked systems

Submission Guidelines

Manuscripts should conform to the standard format as indicated in the "Information for Authors" section of the Paper Submission Guidelines.

All manuscripts to be considered for publication must be submitted by the deadline through Manuscript Central. Select “September 2022/Federated Optimizations” from the drop-down menu of topic titles.

Important Dates

Manuscript Submission Deadline: 1 May 2022 (Firm)
Initial Decision: 1 June 2022
Revised Manuscript Due: 1 July 2022
Final Decision: 15 July 2022
Final Manuscript Due: 1 August, 2022
Publication Date: September/October 2022

Guest Editors

Dan Wang
The Hong Kong Polytechnic University, Hong Kong, China

Zhu Han
University of Houston, USA

Ekram Hossain
University of Manitoba, Canada

Yifei Zhu
Shanghai Jiaotong University, China

Arun Vishwanath
IBM Research, Australia

Choong Seon Hong
Kyung Hee University, Korea

Zhijin Qin
Queen Mary University of London