Deep Learning for Pattern Recognition (DLPR)

Aim and Scopes

Pattern Recognition is one of the most important branches of Artificial Intelligence, which focuses on the description, measurement and classification of patterns involved in various data. In the past 60 years, great progress has been achieved in both the theories and applications of pattern recognition. A typical pattern recognition system is composed of preprocessing, feature extraction, classifier design and postprocessing.

Nowadays, we have entered a new era of big data, which offers both opportunities and challenges to the field of Pattern Recognition. We should seek new Pattern Recognition theories to be adaptive to big data. We should push forward new Pattern Recognition applications benefited from big data.

Deep Learning, which can be treated as the most significant breakthrough in the past 10 years in the field of pattern recognition and machine learning, has greatly affected the methodology of related fields like computer vision and achieved terrific progress in both academy and industry. It can be seen as a resolution to change the whole pattern recognition system. It achieved an endtoend pattern recognition, merging previous steps of preprocessing, feature extraction, classifier design and postprocessing.

It is expected that the development of deep learning theories and applications would further influence the field of pattern recognition.

This special issue mainly focuses on Deep Learning for Pattern Recognition (DLPR). We are soliciting original contributions, of leading researchers and practitioners from academia as well as industry, which address a wide range of theoretical and application issues in deep learning for pattern recognition.

Original papers to survey the recent progress in this exciting area and highlight potential solutions to

common challenging problems are also welcome. The topics include, but not limited to:

• Deep learning architecture for pattern recognition
• Optimization for deep learning
• Supervised deep learning
• Unsupervised deep learning
• Sparse coding in deep learning
• Transfer learning for deep learning
• Deep learning for feature representation
• Deep learning for face analysis
• Deep learning for object recognition
• Deep learning for scene understanding
• Deep learning for text recognition
• Deep learning for dimension reduction
• Deep learning for activity recognition
• Deep learning for biometrics
• Performance evaluation of deep learning

We expect to publish about 12 high quality and topic-related articles in this special issue. Contributions to

this special issue will come both from the attendees of the ICPR 2016 Workshop on Deep Learning for Pattern Recognition, and from other authors working in the field. Of course, in case of submissions that are extended versions of previous conference papers, at least 1/3 new content should be added for the journal version and there should not be parts of the article that are verbatim the same as in the conference paper. In any case, all submitted articles should not have been previously published and should not be submitted for publication elsewhere at the time of the submission to the special issue. A review article (RA) authored by outstanding researchers in the field will be invited in this special issue.

Review process

The review process will be done by following the standard review process of Pattern Recognition Letters, according to which each paper will be reviewed by at least two experts in the field. In general, only two reviewing rounds will be possible, out of which major revision is possible only for the first round. Papers that after the 2nd reviewing round still need major revision will be most possibly rejected.

Submission instructions

Prospective authors are invited to upload their manuscripts during the submissions period. Papers should be prepared by adhering to the PRLetters guidelines, in particular as regards the maximum number of allotted pages (10 pages in the PRL layout). When uploading their papers through the online system,

Authors should select the acronym “SI: DLPR” to make it clear that they are submitting to this special issue.

Guest editors will make an initial determination of the suitability and scope of all submissions.

Important dates (tentative)

Submission period: Feb 15 – Mar 15, 2017
First notification: May 30, 2017
Revised manuscript (for second review) due: June 30, 2017
Acceptance Notification: August 30, 2017
Publication of the special issue: 2018

Guest Editor

Dr. Zhaoxiang Zhang (zhaoxiang.zhang@ia.ac.cn) (Managing Guest Editor)
Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China

Dr. Shiguang Shan (sgshan@ict.ac.cn)
Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China

Dr. Yi Fang (yfang@nyu.edu)
Department of Electrical and Computer Engineering, NYU Abu Dhabi, UAE and NYU Tandon School of
Engineering, USA

Dr. Ling Shao (ling.shao@ieee.org)
School of Computing Sciences, University of East Anglia, Norwich, England