Edge computing has been pushing computing services from clouds toward ends in recent years. Meanwhile, big data on edges are growing with privacy awareness, energy and efficiency concern to compose an urgent need to push the artificial intelligence (AI) applications towards network edges. These two emerging trends led to an interdisciplinary area, i.e. Edge Intelligence (EI), which starts to draw tremendous amounts of academic and industrial attention, even if the research on Edge Intelligence is in its early stage. One category of Edge Intelligence approaches is to make the traditional cloud-based AI models tailored for edges by model compression, model partition, edge caching etc. The other category is to implement the procedure of building AI models entirely on the edge by federated learning, gradient compression, gossip learning etc. Nevertheless, designing EI solutions has to make trade-offs between various aspects of the hardware capability, performance, cost, privacy, efficiency, reliability and be optimized with respect to its specific objectives. This brings many challenging research issues and exchanging latest advances in Edge Intelligence is desired by researchers in all related domains, e.g. computer architectures, computer systems, artificial intelligence, computer network, AI circuits and systems. This special issue aims at the latest and novel contributions from both academic researchers and industry practitioner in the emerging area of Edge Intelligence.
Topics of interest to the Special Issue include but not limited to:
Edge intelligence paradigm for manufacturer pipelines
Edge intelligence paradigm for smart communities
Edge intelligence paradigm for scientific apparatus
Edge intelligence architecture, circuits and systems
Federated learning algorithms and training schemes for edge intelligence systems
Privacy and safety protection for edge intelligence systems
Collaborative management for edge intelligence systems and clouds
Energy management of edge intelligence systems
Sensing and communication technologies for edge intelligence systems
All original manuscripts that fit within the scope are welcome. Authors of papers presented at the SMARTCLOUD series conference are especially encouraged to submit to this SI.
General information for submitting papers to JSA can be found at https://www.journals.elsevier.com/journal-of-systems-architecture. Submissions should be made online at https://www.editorialmanager.com/jsa/. Please select the “VSI:EISA20” option as type of the paper. Please direct all enquiries regarding this SI to the leading guest editor.
JSA has adopted the Virtual Special Issue model to speed up the publication process, where Special Issue papers are published in regular issues, but marked as SI papers. Acceptance decisions are made on a rolling basis.
Therefore, authors are encouraged to submit papers early, and need not wait until the submission deadline.
Submission of manuscripts: 30 Nov. 2020
Notification to authors: 20 Feb. 2021
Prof. Yongxin Zhu, Chinese Academy of Sciences, China
Prof. Meikang Qiu, Texas A&M University, U.S.A.
Asst. Prof. Alireza Jolfaei, Macquarie University, Australia