Secure and Credible Neural Network in Mobile Computing
1Chinese Academy of Sciences, Beijing, China
2Technische Universitat Wien, Wien, Austria
3Amity University, Noida, India
Secure and Credible Neural Network in Mobile Computing
Description
With the rapid adoption of mobile devices, consumer devices, drones, and vehicles, mobile computing has emerged as a field with great impact, potential, and growth. Recently, neural networks, especially deep neural networks (DNNs) have arisen as one of the major drivers in advancing artificial intelligence and have led to vast progress in the field of wireless communication and mobile computing.
With the rapid development of deep learning technology in the field of wireless communication and mobile computing, researchers have developed many powerful DNNs, that depend on large amounts of labeled data. However, due to their inherent black-box characteristics, traditional DNNs are opaque, and can even be unable to explain what they have learned, why they emulated a certain task, or why they make a particular decision as a result. Therefore, developing secure, credible, and explainable neural networks in mobile computing has become a top priority and requires innovative solutions. At the same time, the implementation of DNNs requires powerful computing resources. In the mobile computing environment, mobile devices have limited computing, storage, and network capacity. The application of DNNs suffers from the issues of unstable communications, and volatile mobile networks. Without properly addressing these issues, the wider application of DNNs in practice will be limited, and as such the effective development and efficient execution of secure and credible DNN models in the mobile computing environment has become the focus in academia and industry.
Therefore, this Special Issue seeks research on secure and credible neural networks in mobile computing. This Special Issue will provide researchers with new ideas to aid in solving practical application problems in mobile computing. The goal of this special issue is to present state-of-the-art and high-quality original research and review papers focusing on architecture, algorithms, optimization, tools, and models of secure and credible neural networks in mobile computing.
Potential topics include but are not limited to the following:
- DNN security in mobile computing
- DNN interpretability in mobile computing
- DNN compression in mobile computing
- Security and privacy in mobile computing
- Efficient and secure pattern recognition in mobile computing
- Efficient and secure image processing in mobile computing
- Algorithms, schemes, and techniques of DNN application in mobile computing
- DNN acceleration in mobile computing
- DNN offloading and split learning