Title: Meta-Networking: Overcoming the Shannon Limit with Multi-faceted Information
This talk introduces Meta-networking: a revolutionary networking architecture that provides a beyond-Shannon paradigm with multi-faceted information based on the collaboration among distributed network entities, efficient service classification, and data feature extraction. An overview of Meta-Networking is provided and the key principles and components of Meta-Networking, including the quality-of-experience characterization, AI-empowered semantic encoding, and information density improvement, are analyzed. It enables a groundbreaking communication system where a much larger amount of information is transmitted without increasing the size of binary digits. Furthermore, an application scenario for image transmission in the Internet of Vehicles is discussed, which shows significant performance improvement compared with conventional communications. It is believed that Meta-Networking has the potential for revolutionizing communication systems with higher efficiency, stronger reliability, and intelligence awareness.
Celimuge Wu received his PhD degree from The University of Electro-Communications, Japan. He is currently a professor and the director of Meta-Networking Research Center, The University of Electro-Communications. His research interests include Vehicular Networks, Edge Computing, IoT, and AI for Wireless Networking and Computing. He serves as an associate editor of IEEE Transactions on Cognitive Communications and Networking, IEEE Transactions on Network Science and Engineering, and IEEE Transactions on Green Communications and Networking. He is Vice Chair (Asia Pacific) of IEEE Technical Committee on Big Data (TCBD). He is a recipient of 2021 IEEE Communications Society Outstanding Paper Award, 2021 IEEE Internet of Things Journal Best Paper Award, IEEE Computer Society 2020 Best Paper Award and IEEE Computer Society 2019 Best Paper Award Runner-Up. He is an IEEE Vehicular Technology Society Distinguished Lecturer.
Title: Detecting Anomalies from Big Data System Logs
Parallel and distributed systems play a prominent role due to the large amount of collected data. Although they are effective in many aspects, it is not easy to conduct maintenance and management, thus may cause some severe system problems. To enhance these applications, a range of troubleshooting methods are proposed. Among these methods, log analysis is very popular due to its effective resolution and rich information in events or states of the system, which improves system health diagnosis with root cause analysis. Generally, log-based anomaly detection methods are developed by mining a large set of system log data. The data mining process is usually realized with different models, such as statistical models and non-deep machine learning models. However, these traditional models still confront with low detection accuracy due to focusing on ad-hoc features. Nowadays, deep neural networks (DNNs) have been widely employed to solve log anomaly detection and outperform a range of conventional methods. They have attained such striking success because they can usually explore and extract semantic information from a large volume of log data, which helps to infer complex log anomaly patterns more accurately. Hence, more anomaly detection models are equipped with DNN and break through the bottleneck in the performance.
Siyang Lu is an assistant professor in the School of Computer and Information Technology at the Beijing Jiaotong University. He received Ph.D. in Computer Science from University of Central Florida in 2019. He received his Master Degree in Software Engineering from Tianjin University in 2015. His research focuses on big data computing and abnormal detection techniques in the following aspects: leveraging deep learning techniques to detect and prevent programming errors and execution anomaly in big data and/or parallel programs; improving accuracy and security of log anomaly detection models; optimizing performance and scalability of big data processing and/or parallel computing systems. He received two Best Paper Awards (IEEE CyberSciTech 2017 and IEEE UIC 2022). He is a member of CCF, IEEE, ACM, and Technical Committee on Software Engineering of CCF.
Title: Some Economic Issues and Our Solutions for Edge Computing
In recent years, new Internet technologies such as NDN and IoT have emerged, and the existing economic structure of the Internet no longer fits the new Internet architecture. To solve this problem, this study examined several typical Internet usage scenarios and designed new economic operation models for these scenarios using game theory and auction theory. We believe that these new economic theories and methods will greatly contribute to the future prosperity and development of the Internet industry.
Xun Shao received his Ph.D. in information science from the Graduate School of Information Science and Technology, Osaka University, Japan, in 2013. From 2013 to 2017, he was a researcher with the National Institute of Information and Communications Technology (NICT) in Japan. From 2018 to 2022, he was an Assistant Professor at the School of Regional Innovation and Social Design Engineering, Kitami Institute of Technology, Japan. He is currently an Associate Professor with the Department of Electrical and Electronic Information Engineering, Tohohashi University of Technology, Japan. His research interests include distributed systems and information networking. He is a member of the IEEE and IEICE.