Deep learning has been a popular research area in artificial intelligence with many applications in various of fields, such as object detection and recognition, and multimedia data understanding, restoration, and synthesis. Moreover, deep learning has also played a critical role in data science, especially for analyzing big data relying on the extractions of high-level abstractions for data representations through a hierarchical learning process. In realizing deep learning frameworks, both supervised and unsupervised approaches for training deep neural networks have been empirically investigated and applied in different applications. However, there is still very limited understanding on why deep networks can work well and how to design efficient deep models in the viewpoints of software and hardware. Moreover, for embedding deep models into mobile applications, efficient representation or compression of model parameters for model storage is also highly required.
This special issue will focus on all aspects of deep learning in representation, interpretation, and applications. The special issue is mainly extended from the special session on Recent Advances in Deep Learning with Multimedia Applications of APSIPA ASC 2020 conference, but any other significant contributions in the related fields are also welcome.
Topics of interest include, but are not limited to:
- Interpreting and Understanding Deep Neural Networks
- Deep Model Compression and Representation
- Deep Representation Learning with Multimedia Applications
- Deep Learning-based Multimedia Data Synthesis
- Deep Learning for Big Data Analytics
- Hardware Acceleration for Deep Learning
Editors of the Special Issue
Dr. Li-Wei Kang, National Taiwan Normal University, Taiwan
Dr. Chia-Hung Yeh, National Taiwan Normal University, Taiwan
Submission deadline: June 30, 2021
First Review Decision: August 15, 2021
Revisions Due: September 30, 2021
Final Decision: October 15, 2021
Final Manuscript: November 1, 2021
Expected Publication Date: January 2022
Papers are published upon acceptance, regardless of the Special Issue publication date.
Please refer to the Instructions for Contributors for paper submission procedure.