Artificial Intelligence and Data Science

Innovations in artificial intelligence (AI) and a paradigm shift to data-driven approaches owing to the growing trend in data drive new research opportunities in a variety of areas, such as social networks, bioinformatics, healthcare, manufacturing business, beyond 5G (6G) communications, Internet of Things (IoT), and so forth.  

In the aforementioned areas, system-generated information such as smart devices, sensors, agents, and meters as well as human-generated information such as texts, photos, and videos lead to a tremendous amount of data while new levels of security, performance, and reliability are required.

 In this context, equipping the relevant functionality with AI or data mining-based algorithms, including regression models, Bayesian learning, clustering, neural networks, decision trees, information retrieval, decision processes, multi-armed bandits, reinforcement learning, generative models, and graphical models, has received a substantial attention both in academia as well as in industrial communities. 

Recently developed AI or data mining approaches will provide promising solutions to many challenging problems through learning and decision making in terms of significantly improving user experience and service quality.

 

The “Artificial Intelligence and Data Science” bringing this perspective to AI and data science, and focuses on the latest research, algorithm design, analysis, and implementation for various applications. Will address a comprehensive overview of how to enable autonomous and intelligent services/applications though collecting, processing, learning, and controlling a vast amount of information across various domains. The following topics of interest include, but are not limited to:

 

• Network mining and graph mining

• Deep learning and neural network-based approach

• Social network analysis

• Reinforcement learning and multi-armed bandits 

• Knowledge representation and reasoning

• Anomaly and fake content detection

• Information retrieval

• Recommendation and ranking engines

• Machine learning in medicine and healthcare informatics 

• Big data analytics for beyond 5G or 6G

• Edge/fog computing using machine learning

• IoT data analytics

• Data-driven services and applications

6G Wireless Systems

While 5G is currently being deployed around the globe, research on 6G is under way aiming at addressing the coming challenges of drastic increase of wireless data traffic and support of other usage scenarios. 6G is expected to extend 5G capabilities even further. Higher bitrates (up to Tbps) and lower latency (less than 1ms) will allow introducing new services – such as pervasive edge intelligence, ultra-massive machine-type communications, extremely reliable low-latency communications, holographic rendering and high-precision communications – and meet more stringent requirements, especially in the following dimensions: energy efficiency; intelligence; spectral efficiency; security, secrecy and privacy; affordability; and customization. Artificial intelligence approaches and techniques, such as machine learning (of which deep learning and reinforcement learning are specific examples), and machine reasoning (which includes planning, scheduling, knowledge representation and reasoning, search and optimization), are the new fundamental enablers to operate networks more efficiently, enhance the overall end user experience and provide innovative service applications. Quantum Optics Computing (QOC) and Quantum Key Distribution (QKD) are almost ready for industrial applications. In particular, massive Internet of Things (mIoT), Industrial IoT (IIoT), fully automated robotic platforms (which include control, perception, sensors and actuators, as well as the integration of other techniques into cyber-physical systems), vehicles and multisensory extended reality are examples of the new data-demanding applications, which will impose new performance targets and motivate 6G design and deployment.

This paper aims to provide the scientific community with a comprehensive overview of the most challenging aspects of 6G mobile networks and identify latest research on promising techniques towards the evolution to 6G on topics including, but not limited to the following:

- Vision, key drivers, new services and requirements for 6G

- System and network architectures for 6G 

- Wireless backhaul and fronthaul solutions

- Spectrum and channel modeling for 5G and towards 6G

- Energy efficiency and harvesting technologies

- Multi-level machine learning pipelines in 5G and towards 6G

- 5G and beyond towards 6G testbeds and experimentation

- Security, secrecy and privacy schemes for 5G and towards 6G

- Distributed computing for 5G and towards 6G

- Optical Quantum computing and QKD in 6G