Speaker: Professor Dusit Niyato, Ph.D., IEEE Fellow and IET Fellow, School of Computer Science and Engineering, Nanyang Technological University, Singapore

Host: Professor Xiang Cheng, Ph.D., Boya Distinguished Professor, IEEE Fellow, from State Key Laboratory of Advanced Optical Communication Systems and Networks, School of Electronics, Peking University

Lecture 1

Theme 1: Semantic Communication Meets Edge Intelligence
Theme 2: Distributed Machine Learning for 6G Networks

Time: 2:00-4:00 p.m., August 23, 2022, GMT+8

Venue: Tecent Meeting ID: 911-498-527

Lecture 2

Theme 3: Metaverse Virtual Service Management: Game Theoretic Approaches
Theme 4: Borrowing Arrows with Thatched Boats: The Art of Defeating Reactive Jammers in IoT Networks

Time: 3:00-5:00 p.m., August 26, 2022, GMT+8

Venue: Tecent Meeting ID: 791-958-810

Lecture 3

Theme 5: Resource Allocation in Coded Distributed Computing
Theme 6: Deep Reinforcement Learning and Its Applications for Future Wireless Networks

Language: English (without interpretation)

Abstract:

Theme 1: The development of emerging applications, such as autonomous transportation systems, is expected to result in an explosive growth in mobile data traffic. As the available spectrum resources become more and more scarce, there is a growing need for a paradigm shift from Shannon’s classical information theory to semantic communication. Specifically, the former adopts a “transmit-before-understanding” approach while the latter leverages artificial intelligence techniques to “understand-before-transmit”, thereby alleviating bandwidth pressure by reducing the amount of data to be exchanged without negating the semantic effectiveness of the transmitted symbols. In this presentation, we first introduce background and present the framework of the modern semantic communication. Then, we discuss the general semantic extraction methods and semantic metrics. To reduce the computation and storage overheads incurred by the semantic extraction procedure, we introduce an edge-driven training, maintenance, and execution of semantic extraction. We further investigate how edge intelligence can be enhanced with semantic communication by improving the generalization capabilities of intelligent agents at lower computation overheads and reducing the communication overhead of information exchange. Finally, we present case studies involving semantic-aware resource optimization for the wireless internet of things and attention-based reinforcement learning for performance optimization for semantic communications.

Theme 2: As the successor of 5G, 6G technologies are expected to bring breakthroughs as well as a comprehensive revolution for future mobile networks. Specifically, it is expected to support applications far beyond anything seen so far, such as holographic communication-based extended reality and tactile Internet. Therefore, some strict requirements have been put in place for 6G networks, for example, super-high data rate (e.g., up to 1Tb/s), very broad frequency-bands (e.g., 73GHz-140GHz and 1THz-3THz), and less than 1-millisecond end-to-end latency. To meet such requirements, distributed machine learning has been considered as a core component, which can enable 6G networks to achieve breakthroughs in communication technologies. Specifically, distributed machine learning techniques, e.g., federated learning, split learning, distributed transfer learning, and multi-agent reinforcement learning, can be implemented to design and optimize 6G architecture and network operations in a cost-efficient manner. In addition, through a huge amount of data collected from mobile applications, mobile users as well as network operators, distributed machine learning approaches allow to synthesize, analyze and provide important information for optimizing operations, thereby meeting stringent requirements for security, speed, and performance of 6G networks. Therefore, the main objectives of this tutorial are to provide an overview of distributed machine learning techniques and then study their recent advances to address practical challenges in 6G networks.

Theme 3: Metaverse is the next-generation Internet after the web and the mobile network revolutions, in which humans (acting as digital avatars) can interact with other people and software applications in a three-dimensional (3D) virtual world. In this lecture, we first briefly introduce major concepts of Metaverse and the virtual service management. Then, we discuss applications of game theory in the virtual service management. First, we consider that virtual reality (VR) users in the wireless edge-empowered Metaverse can immerse themselves in the virtual through the access of VR services offered by different providers. The VR service providers (SPs) have to optimize the VR service delivery efficiently and economically give their limited communication and computation resources. An incentive mechanism can be thus applied as an effective tool for managing VR services between providers and users. Therefore, we propose a learning-based incentive mechanism framework for VR services in the Metaverse. Second, we consider virtual services provided through the digital twin, i.e., a digital replication of real-world entities in the Metaverse. The real-world data collected by IoT devices and sensors are key for synchronizing the two worlds. A group of IoT devices are employed by the Metaverse platform to collect such data on behalf of virtual service providers (VSPs). Device owners, who are self-interested, dynamically select a VSP to maximize rewards. We adopt hybrid evolutionary dynamics, in which heterogeneous device owner populations can employ different revision protocols to update their strategies. To this end, we discuss some important research directions in Metaverse virtual service management.

Theme 4: In this lecture, we introduce a novel deception strategy, which is inspired by the "Borrowing Arrows with Thatched Boats'' strategy, one of the most famous military tactics in the history noted in historical fiction series A Romance of Three Kingdoms. The purpose is to defeat reactive jamming attacks for low-power Internet-of-Things (IoT) networks. Our proposed strategy allows resource-constrained IoT devices to be able to defeat powerful reactive jammers by leveraging their own jamming signals. More specifically, by stimulating the jammer to attack the channel through transmitting fake transmissions, the IoT system can not only undermine the jammer's power, but also harvest energy or utilize jamming signals as a communication means to transmit data through using Radio Frequency (RF) energy harvesting and ambient backscatter techniques, respectively. Furthermore, we develop a low-cost deep reinforcement learning framework that enables the hardware-constrained IoT device to quickly obtain an optimal defense policy without requiring any information about the jammer in advance. Simulation results reveal that our proposed framework can not only be very effective in defeating reactive jamming attacks, but also leverage jammer's power to enhance system performance for the IoT network.

Theme 5: Due to benefits such as high reliability, scalability, computation speed, and cost-effectiveness, distributed computing has become a common approach for large-scale computation tasks. However, distributed computing faces critical issues related to communication load and straggler effects. In particular, computing nodes need to exchange intermediate results with each other in order to calculate the final result, and this significantly increases communication overheads. Furthermore, a distributed computing network may include straggling nodes that run intermittently slower. This results in a longer overall time needed to execute the computation tasks, thereby limiting the performance of distributed computing. To address these issues, coded distributed computing (CDC), i.e., a combination of coding theoretic techniques and distributed computing, has been recently proposed as a promising solution. Coding techniques have proved effective in WiFi and cellular systems to deal with channel noise. Therefore, CDC may significantly reduce communication load, alleviate the effects of stragglers, provide fault-tolerance, privacy and security. In this lecture, we will discuss the motivation of the research of CDC, review several CDC works and present the applications of CDC schemes. We also discuss one of our recent works that focuses on the design of an incentive mechanism for the effective implementation of scalable and efficient CDC schemes over the distributed computing network.

Theme 6: Modern networks, e.g., Internet of Things (IoT) and unmanned aerial vehicle (UAV) networks, become more decentralized and autonomous. In such networks, network entities need to make decisions locally to maximize the network performance under uncertainty of network environment. Reinforcement learning has been efficiently used to enable the network entities to obtain the optimal policy including, e.g., decisions or actions, given their states when the state and action spaces are small. However, in complex and largescale networks, the state and action spaces are usually large, and the reinforcement learning may not be able to find the optimal policy in a reasonable time. Therefore, deep reinforcement learning (DRL), a combination of reinforcement learning with deep learning, has been developed to overcome the shortcomings. In this lecture, we aim to provide a fundamental background of DRL and then study recent advances in DRL to address practical challenges in wireless networks. In particular, we first give a tutorial of deep reinforcement learning from basic concepts to advanced models to motivate and provide fundamental knowledge for the audiences. We then provide a case study together with implementation details to help the audiences to have a good understanding of how to practice with DRL. After that, we review DRL approaches proposed to address emerging issues in communications and networking. Finally, we highlight important challenges, open issues, and future research directions of applying deep reinforcement learning.

Biography:

Dusit Niyato is currently a professor in the School of Computer Science and Engineering, Nanyang Technological University, Singapore. He received B.E. from King Mongkuk’s Institute of Technology Ladkrabang (KMITL), Thailand in 1999 and Ph.D. in Electrical and Computer Engineering from the University of Manitoba, Canada in 2008. Dusit's research interests are in the areas of distributed collaborative machine learning, Internet of Things (IoT), edge intelligent metaverse, mobile and distributed computing, and wireless networks. Dusit won the Best Young Researcher Award of IEEE Communications Society (ComSoc) Asia Pacific and The 2011 IEEE Communications Society Fred W. Ellersick Prize Paper Award and the IEEE Computer Society Middle Career Researcher Award for Excellence in Scalable Computing in 2021 and Distinguished Technical Achievement Recognition Award of IEEE ComSoc Technical Committee on Green Communications and Computing 2022. Dusit also won a number of best paper awards including IEEE Wireless Communications and Networking Conference (WCNC), IEEE International Conference on Communications (ICC), IEEE ComSoc Communication Systems Integration and Modelling Technical Committee and IEEE ComSoc Signal Processing and Computing for Communications Technical Committee 2021. Currently, Dusit is serving as Editor-in-Chief of IEEE Communications Surveys and Tutorials, an area editor of IEEE Transactions on Vehicular Technology, editor of IEEE Transactions on Wireless Communications, associate editor of IEEE Internet of Things Journal, IEEE Transactions on Mobile Computing, IEEE Wireless Communications, IEEE Network, and ACM Computing Surveys. He was a guest editor of IEEE Journal on Selected Areas on Communications. He was a Distinguished Lecturer of the IEEE Communications Society for 2016-2017. He was named the 2017-2021 highly cited researcher in computer science. He is a Fellow of IEEE and a Fellow of IET.

Source: PKU_IFC