FOREWORD

Chi-Hua CHEN

Summary :

• Publication: IEICE TRANSACTIONS on Information Vol.E106-D No.5 pp.579-580
• Publication Date: 2023/05/01
• Online ISSN: 1745-1361
• DOI: 10.1587/transinf.2022DLF0001
• Type of Manuscript: FOREWORD

Special Section on Deep Learning Technologies: Architecture, Optimization, Techniques, and Applications

Deep learning techniques, including neural network (NN), convolutional neural network (CNN), graph neural network (GNN), graph convolutional network (GCN), recurrent neural network (RNN), long short-term memory (LSTM) network, gated recurrent unit (GRU) network, have been popularly applied to data analyses and management. For instance, the CNN and auto-encoder can be used to analyze the pattern recognition and extract the features of data in various applications such as regression, classification, image recognition, etc. Furthermore, the RNN, LSTM network, and GRU network can be used to perform the time-series inference for chronology-oriented data (e.g. speech data, weather data, transportation data, stock market data, etc.). Regarding the applications in transportation, the advanced driver assistance systems and autonomous cars have been extensively developed based on the deep learning techniques, which enable the forward collision warning, blind spot monitoring, lane departure warning, traffic sign recognition, traffic safety, infrastructure management and congestion, and so on. However, how to enhance the performance and efficiency of these deep learning techniques is one of the biggest challenges of implementing these real-time applications.

Several optimization techniques, e.g. stochastic gradient descent (SGD), adaptive moment estimation (Adam), Nesterov-accelerated adaptive moment estimation (Nadam) algorithms have been proposed to support the deep learning algorithms for efficient solution-searching. For instance, the gradient descent method is a popular optimization technique to quickly seek the optimized weight sets and filters of the CNN for image recognition. Moreover, the hybrid approaches typical of mathematics in engineering and computer science such as the deep learning and optimization techniques can be investigated and developed to support a variety of data analyses and management.

Topics covered in this issue are categorized into eight themes: (1) core methods, (2) smart industry, (3) smart healthcare, (4) smart agriculture, (5) intelligent transportation systems, (6) positioning and navigation, (7) object recognition and tracking, and (8) person image generation. This special issue receives a total of 89 submissions, only 28 articles of which are accepted. An acceptance rate of 31.46% stems from the rigid review process for ensuring that only quality research work with significant results are collected. Relevant statistics of the special issue are shown as follows:

• 89 submissions;
• 28 publications: 6 letters and 22 papers;
• 61 rejections/withdrawals.

We sincerely thank the Institute of Electronics, Information and Communication Engineers (IEICE) for their great support. We are also grateful to all the editors and reviewer, as well as all the authors, for their substantial contributions and efforts.

Special Section Editorial Committee Members

Guest Editor-in-Chief:

Abel C. H. Chen (Chunghwa Telecom Co., Ltd.)

Guest Associate Guest Editor-in-Chiefs:

Wenzhong Guo (Fuzhou University)

Yi-Bing Lin (National Yang Ming Chiao Tung University)

Haishuai Wang (Zhejiang University)

Kuo-Ming Chao (Bournemouth University)

Feng-Jang Hwang (National Sun Yat-sen University)

Cheng Shi (Xi’an University of Technology)

Hsu-Yang Kung (National Pingtung University of Science and Technology)

Guest Associate Editors:

Shaoen Wu (Illinois State University)

Xianbiao Hu (Missouri University of Science and Technology)

Xiao-Guang Yue (European University Cyprus)

Hanhua Chen (Huazhong University of Science and Technology)

Yuh-Shyan Chen (National Taipei University)

Shui Yu (University of Technology Sydney)

Ting Bi (Maynooth University)

Tianxi Dong (Trinity University) Chunjia Han (Birkbeck, University of London)

Mu-Yen Chen (National Cheng Kung University)

Xiongbiao Luo (Xiamen University)

Nianyin Zeng (Xiamen University)

Wen-Kang Jia (Fujian Normal University)

Jiayuan Xin (Newcastle University)

Usman Tariq (Prince Sattam bin Abdulaziz University)

Moayad Aloqaily (Gnowit Inc.)

K. Shankar (South Ural State University)

Paula Fraga-Lamas (Universidade da Coruña)

Meng-Hsun Tsai (National Cheng Kung University)

Fuquan Zhang (Minjiang University)

Chih-Min Yu (Yango University)

Chin-Ling Chen (Chaoyang University of Technology)

Hao-Chun Lu (National Yang Ming Chiao Tung University)

Yao-Huei Huang (Fu-Jen Catholic University)

Hao-Hsiang Ku (National Taiwan Ocean University)

Mingyang Pan (Dalian Maritime University)

Fangying Song (Fuzhou University)

Genggeng Liu (Fuzhou University)

Yu-Chih Wei (National Taipei University of Technology)

Hsiao-Ting Tseng (National Central University)

Chia-Yu Lin (National Central University)

Authors