Parallel Deep Neural Networks: Theory, Methods and Applications
A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".
Deadline for manuscript submissions: closed (20 August 2023) | Viewed by 251
Special Issue Editor
Special Issue Information
Dear Colleagues,
Deep neural networks are both computation- and data-intensive, posing key challenges during both inference and training phases. On one hand, model inference is expected to be deployed onto mobile platforms with restricted power and form factor, but with real-time performance requirements. On the other hand, with the increase of the size of datasets, the multi-modality of data, and the complexity of models, algorithmic/software optimization for training on high-performance general-purpose architectures as well as hardware acceleration through specialized architectures are also critically important. Furthermore, as research in this field progresses, the focus is on multi-purpose network architectures adapted to a wide range of downstream tasks, with billions of parameters, that are becoming deeper and more interconnected. This situation will become even more important with new advances coming at an increasing pace, creating unexplored opportunities for parallelization and programming frameworks to design parallel and distributed algorithm architectures.
This Special Issue is dedicated to cutting-edge research and recent advances in the field of parallel deep neural networks; both theoretical and experimental studies are welcome, including topics such as parallelism strategies and techniques to reduce communication, training on distributed environments, highly energy-efficient heterogeneous computing platforms, hardware-efficient training and inference, and application performance modeling.
Dr. Cesar Torres-Huitzil
Guest Editor
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
Keywords
- deep neural networks
- data/model parallelism
- high-performance architectures
- hardware accelerators
- performance modeling
- scalable optimization
- neural architecture search
