Edge / Cloud Computing Meets Artificial Intelligence

Summary and Scope

Recent years have witnessed the proliferation of mobile computing and Internet-of-Things (IoT), where billions of mobile and IoT devices are connected to the Internet, generating zillions bytes of data at the network edge. Edge-Cloud Computing, a continuously emerging parallel & distributed computing paradigm, has received a tremendous amount of attention. By pushing data storage, computing, and controls closer to the network edge, edge computing has been widely recognized as a promising solution to meet the requirements of low latency, high scalability and energy efficiency. Owing to the recent developments in the domain of neural networks and cloud computing, Artificial Intelligence (AI) has been applied to a variety of disciplines and proved highly successful in a vast class of intelligent applications across many domains, e.g., computer vision, pattern recognition, etc.

Recently, edge intelligence, aiming to facilitate the deployment of neural networks on edge computing, has received significant attention, since hierarchical architecture of end devices proposes a possible solution to meet the high computation and low-latency requirement for the training and inference of AI algorithms. However, there are many challenges existing for novel designs of edge-cloud computing architectures for AI applications, and their co-optimization. On one hand, the high resource requirements of AI applications should be accommodated on a set of less powerful edge compute resources. Therefore, efficient, parallel & distributed and resource-conserving (e.g., memory, computing, energy, time, etc.) AI algorithms should be revisited in the edge computing environments and move the main part to cloud computing. On the other hand, the system design should also support the efficient and scalable execution of AI algorithms, including specialized accelerators, efficient parallel & distributed execution mode, optimal off-loading, and scheduling strategies, etc.

In this special issue, we solicit original work exclusively on ML/AI, specifically catered to deep neural networks on/for edge computing and efficient learning systems or accelerators on edge computing, addressing specific challenges in this field. The list of possible topics includes, but not limited to:

• Parallel & distributed computing architectures for edge-cloud based AI

• Edge-Cloud collaborative computing for neural networks

• Edge/Fog-infused cloud architectures for ML/AI applications

• Power-aware efficient ML/AI algorithms for edge devices

• Parallel & distributed neural networks for edge/fog/cloud computing

• Offloading & scheduling strategy for edge AI

• Osmotic and catalytic computing strategies for edge/fog/cloud platforms

• Data parallelism and model parallelism on edge/fog/cloud computing

• Hardware-aware ML/AI algorithms on edge/fog/cloud computing

• Few-shot learning on edge devices for ML/AI applications

• Resource scheduling for large-scale applications of edge intelligence

• AI/ML algorithms for small-scale low-power edge devices

• AI/ML algorithms for edge/fog/cloud platforms

• Distributed and cooperative learning with edge devices on Cloud

• Architecture & applications of edge AI for IoT

• Accelerators for edge-cloud AI

Submission Guidelines

​Authors should follow the Journal of Parallel and Distributed Computing manuscript format described at the journal site: https://www.elsevier.com/journals/journal-of-parallel-and-distributed-computing/0743-7315/guide-for-authors. Manuscripts that extend research published previously (e.g., in conference or workshop proceedings) will only be considered if they include at least 30% of significantly new material; the submission of such manuscripts must be accompanied by a “Summary of Differences” letter explaining how the authors extended their previously published work. All manuscripts and any supplementary material should be submitted through Editorial Manager (EM), available at: https://www.editorialmanager.com/jpdc/default.aspx. The authors must select “VSI: Edge/Cloud Meets AI” when they reach the “Article Type” in the submission process.

Important Dates

Final Submission Date: 1 September 2021

Final Acceptance Date: 1 June 2022

Guest Editors

Prof. Bharadwaj Veeravalli

Department of ECE,

Faculty of Engineering, NUS, Singapore

Email: [email protected]

Prof. Zeng Zeng

Institute for Infocomm Research, Agency for Science, Technology and Research (A*STAR), Singapore

Email: [email protected]

Prof. Keqin Li

IEEE Fellow

SUNY Distinguished Professor, USA

Email: [email protected]