Deep Learning for the Internet of Things

Dear Colleagues,

The recent advancements in artificial intelligence; machine learning; and, more importantly, deep learning, with its many variations of algorithms and platforms, have radically transformed conventional software applications into a set of smart and connected components with the capability of data-driven decision-making; intelligent data sensing and filtering; proactive and collaborative integration of software and hardware devices; and, more importantly, the ability to capture tons of useful interaction data. Machine and deep learning-based components are becoming the backbone of modern systems such as cyber-physical systems (CPS) and Internet of Things (IoT), where intelligent integrations and collaborations of smart devices along with their outstanding data collection capabilities make them a perfect application domain for such sophisticated learning algorithms. The beauty of smart and connected devices governed by software utilities has already been demonstrated in several application domains such as healthcare, agriculture and farming, manufacturing, smart buildings, transportations, energy, and environmental surveillance monitoring systems. While these IoT-based systems can capture a good amount of data, the usage of these valuable and hard-to-produce data are not being fully utilized. Deep learning enables one to explore these data in further detail and capturing the hidden relationships that may exist amongst their key factors and parameters, thus providing further insight into the underlying application domains. Furthermore, deep learning algorithms are also capable of addressing challenging problems that are natural in IoT but remain unsolved in systems such as real time optimization problems, modeling non-linear characteristics of IoT devices and components, and several instances of prediction and classification problems.

This Special Issue aims to foster deep learning-based modeling, solutions, and approaches to problems in Internet of Things systems. It seeks to explore deep learning algorithms, including generative adversarial models, attention-based networks, deep reinforcement learning, and recurrent deep neural networks, in capturing features and modeling the behavior of the involved software and hardware components.

Academic researchers and industrial practitioners are invited to contribute their valuable research and experiences to state-of-the-art of deep learning-based approaches to modeling Internet of Things and associated problems. Topics of interest include but are not limited to the following applications of deep learning techniques to

• Modeling IoT systems using deep learning;
• Generative adversarial networks (GANS) in IoT and CPS;
• Long short-term memory (LSTM) modeling of IoT time series data;
• Attention-based approaches to capture significant features in IoT;
• Deep learning-based modeling and experience in IoT-based applications such as smart building, healthcare, agriculture, manufacturing, left-driving cars, and cyber security;
• Deep reinforcement learning for modeling decision making and uncertainty in IoT.

Prof. Dr. Akbar Siami-Namin
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. Electronics 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.

• deep learning
• Internet of Things
• generative adversarial networks
• attention network