Deep Learning-based Intelligent Systems: Theories, Algorithms, and Applications (SI-dlis)

Overview

Deep learning has become a topic of increasing interest for researchers, from both academia and Industry, during the past decade. Unlike traditional machine learning methods, deep learning algorithms show the ability to learn and model very large-scale data sets. Deep learning techniques have achieved great success in different tasks in computer vision, natural language processing, robotics, and other areas. Recent years have witnessed a great development of the deep learning theory and various applications in the general field of artificial intelligence, including neural network structure, optimization, data representation, and deep reinforcement learning. Some extensions to the deep learning networks, e.g., attention mechanism, adversarial generative networks, and deep Q-network, were also developed, and remarkable achievements and successes have been achieved.

Although deep learning has been well studied in recent years, there exist many challenges to apply deep learning techniques in intelligent systems. First, deep learning approaches require a huge and diverse amount of data as input to models, and have a large number of parameters for training. Second, the training of deep models is easy to fall into over-fitting problems, and the transfer learning of deep models to other fields is also challenging. Besides, since deep learning models have transparency or black-box issues, it is hard to understand how a given system makes a decision, which, however, is important in some domains such as financial trading or medical diagnosis.

Topics

This special section solicits high-quality papers reporting on deep learning-based intelligent systems, with the goals of highlighting new achievements and developments as well as feature outstanding open issues and promising new directions on theories, algorithms, and applications. Particularly, the principal technical areas could be:

• Knowledge representation, storage, and processing
• Optimization and decision-making
• Learning-based reasoning techniques
• Planning and scheduling
• Cross-modal learning
• Exploring new models and datasets
• Transfer learning
• Deep reinforcement learning
• Attention mechanism
• Adversarial learning
• Multi-modal fusion and knowledge discovery
• Intelligent transportation

Submission Guidelines:

Unpublished manuscripts, or extended versions of papers presented at related conferences, are welcome. Submissions must not be currently under review for publication elsewhere. Conference papers may be submitted only if they are completely re-written or substantially extended (more than 50%), and must be referenced. All submitted papers will be peer-reviewed using the normal standards of the Journal. By submitting a paper to this issue, the authors agree to referee one paper (if asked) within the time frame of the special section.

Before submission, authors should carefully read the Guide for Authors available at
https://www.journals.elsevier.com/computers-and-electrical-engineering

Authors should submit their papers through the journal’s web submission tool at evise.com/profile/#/COMPELECENG/login by selecting “SI-dlis” from the “Issues” pull-down menu during the submission process. For additional questions, contact the Main Guest Editor.

Schedule:

• Submission of manuscript: July 31, 2020
• First notification: August 30, 2020
• Submission of revised manuscript: September 30, 2020
• Notification of the re-review: October 20, 2020
• Final notification: October 31, 2020
• Final paper due: November 15, 2020
• Publication: February 2021

Guest Editors:

• Main Guest Editor: Feiran Huang, Jinan University, China, huangfr@jnu.edu.cn
• Guest Editor: Shahid Mumtaz, Instituto de Telecomunicações, Aveiro, Portugal, smumtaz@av.it.pt

Feiran Huang received his B.Sc. degree from Central South University, Changsha, China, in 2011. He received his Ph.D. degree in computer software and theory from School of Computer Science and Engineering, Beihang University, Beijing, China, in 2019. He is currently an assistant professor at School of Information Science and Technology & College of Cyber Security, Jinan University, Guangzhou, China. He has published nearly 20 papers at top conferences and journals, such as IEEE Transactions on Image Processing, IEEE Transactions on Cybernetics, Knowledge-based Systems, ACM Multimedia Conference (MM) 2018, ACM Multimedia Conference (MM) 2017, The Conference on Information and Knowledge Management (CIKM) 2019, The Conference on Information and Knowledge Management (CIKM) 2018, and International Conference on Multimedia Retrieval (ICMR) 2018. His research interests include deep learning, social media analysis, intelligent systems, and multi-modal learning.

Shahid Mumtaz has more than 10 years of wireless industry/academic experience and is currently working as Senior Research Scientist at Instituto de Telecomunicações Portugal. He has received his Master and Ph.D. degrees in Electrical & Electronic Engineering from Blekinge Institute of Technology, Sweden, and University of Aveiro, Portugal in 2006 and 2011, respectively. Dr. Mumtaz has published four books and published more than 150 publications in very high-rank IEEE transactions, journals, books, book chapters, international conferences. He has organized several workshops as Chair, Special Issues (SI) as Lead Guest Editor of IEEE Communication and Wireless Magazine, and IEEE Transaction on Vehicular Technology. Dr. Mumtaz has been giving invited tutorials/talks in IEEE conferences and been invited to give lectures in different foreign universities. His research interests include vehicular communication, deep learning, and intelligent systems.