Speakers
| Assoc. Prof. Pavel Loskot Zhejiang University-University of Illinois at Urbana-Champaign Institute, ZJUI, China Bio:Pavel Loskot obtained BSc degree in Biomedical Engineering and MSc degree in Radioelectronics from the Czech Technical University of Prague, Czech Republic. He was then Research Scientist and Project Manager in the Centre for Wireless Communications at the University of Oulu, Finland before moving to Canada to earn the PhD degree in Wireless Communications from the University of Alberta. Since 2007, he was a Senior Lecturer in Engineering at Swansea University, UK. He is the Senior Member of the IEEE, Fellow of the Higher Education Academy in the UK, and the Recognized Research Supervisor of the UK Council for Graduate Education. His current research interests lie in applications of statistical signal processing methods to problems in telecommunication engineering and computational molecular biology. Title:Quantifying Uncertainty via Conformal Predictions Abstract:In many scenarios, it is useful to understand how good the estimated or predicted values are, especially when the observations are very noisy. One option is to evaluate the parameter likelihood or even posterior distribution. This may, however, be problematic when more sophisticated machine learning methods such as deep neural networks are used. On the other hand, conformal prediction is a simple and model-agnostic method for obtaining credible or confidence bounds very likely containing the true values. The uncertainty bounds can be also used for other machine learning tasks such as measuring the model uncertainty or deciding how likely it is that the sample comes from a training distribution. In this talk, we will introduce conformal predictions, outline how they are related to quantile regression, then discuss their key statistical properties, and finally explain how conformal predictions can be used in machine learning. |
|
Xi’an Jiaotong University, China Bio:Dr. Hu received the M.Eng. degree in Information and Communications Engineering from Xi'an Jiaotong University, Xi'an, China, in 2016, and the Ph.D. degree in Electronic and Electrical Engineering from University College London (UCL), London, U.K., in 2020. From 2019 to 2021, she was a Research Fellow with the Department of Electronic and Electrical Engineering, UCL, U.K. She is currently a Distinguished Researcher/Associate Professor with the School of Information and Communications Engineering, Xi'an Jiaotong University, Xi'an, China. Her research interests are in the areas of 5G&6G wireless communications, including topics such as edge computing, reconfigurable intelligent surface (RIS)&intelligent reflecting surface (IRS), UAV communications, integrated sensing and communications (ISAC), secure&covert communications, and learning-based communications. She has published one WILEY book chapter on IRS-Aided Mobile Edge Computing: From Optimization to Learning, and 30+ research papers including 2 ESI highly cited papers and 4 Journal Popular Papers. Dr. Hu is an IEEE Member. She has been selected as a member of Young Elite Scientists Sponsorship Program of the China Association for Science and Technology, and a member of the Qin Chuang Yuan High-Level Innovation and Entrepreneurship Talent Program. She has been honored as a “Si Yuan Scholar” sponsored by Huawei Technologies Co., Ltd. Dr. Hu has served as a Guest Editor for ELECTRONICS on Physical Layer Security and for CHINA COMMUNICATIONS Blue Ocean Forum on MAC and Networks. She was a recipient of an Exemplary Reviewer for IEEE COMMUNICATIONS LETTERS. Since 2020, she has been serving as the Assistant to the Editor-in-Chief of IEEE WIRELESS COMMUNICATIONS LETTERS. Title: The applications of RIS/STAR-RIS in wireless communication networks Abstract: RIS has attracted widespread attention from both academic and industry due to its advantages such as low cost, programmability, and easy deployment, which has been considered as one of the key enabling technologies for 6G. In order to further develop and explore the applications of RIS in wireless communication networks, some new RIS structures have been proposed, including simultaneously transmitting and reflecting RIS (STAR-RIS), active RIS, and hybrid RIS, etc. However, the emergence of these new structures also brings new opportunities and challenges. In this report, we first introduce the application potentials of the STAR-RIS in enhancing covert communications as well as the joint implementation of physical layer security (PLS) and covert communications. Then, taking advantage of the 360° coverage range of STAR-RIS, we propose a STAR-RIS-assisted UAV-enabled mobile edge computing (MEC) architecture. Finally, we introduce the solution for integrated sensing and communications (ISAC) scheme assisted by the hybrid RIS. The results show the significant potential of RIS/STAR-RIS in enhancing the performance of wireless communications networks. |