主题演讲人
武庆庆
Associate Professor
上海交通大学
演讲标题:Intelligent Reflecting Surface Empowered 6G Wireless Networks
摘要: Deepfakes are artificially created media posing as actual video recordings and are a potential source of fake news or disinformation. Although research has been done in developing algorithms for the automatic detection of deepfakes, there needs to be more work conducted on how users identify deep fakes. This is a critical missing link because algorithms are currently not performing at a level where human judgement is unneeded. This presentation will discuss the verification strategies users adopt when engaging with content from deepfakes and the implications on businesses and societies.In this talk, we introduce a new wireless research paradigm by employing a massive number of low-cost passive reflecting elements with controllable phase, named intelligent reflecting surface (IRS), which is able to smartly change the wireless signal propagation to enable various functions such as beamforming and interference nulling/cancelation. We illustrate the main applications of IRS in achieving spectrum and energy efficient as well as secure and sustainable wireless networks in the future, and its advantages as compared to existing technologies such as massive MIMO and active relaying. We then present the signal and channel model of IRS by taking into account its hardware limitation in practice. Next, we focus on discussing the main design challenges in IRS-aided wireless networks, including joint active and passive beamforming optimization, channel estimation, etc., and highlight important directions for sensing, powering and computing.
简历:武庆庆，上海交通大学，副教授。他目前的研究兴趣包括智能反射表面（IRS）、无人机通信和MIMO收发器设计。他与他人合著了100多篇IEEE期刊论文，其中包括30多篇被ESI高度引用论文和9篇ESI热门论文，共获得了超过26,000次谷歌引用。他分别于2022年和2021年被列为Clarivate ESI高被引研究员，2021年获得Aminer颁发的AI-2000最具影响力学者奖，并于2020年和2021年被斯坦福大学评为世界顶尖2%的科学家。
他曾获得IEEE通信学会Fred Ellersick奖、2023年IEEE最佳教程论文奖、2022年亚太地区最佳青年研究员奖和优秀论文奖、2021年年轻作者最佳论文奖、2017年中国通信学会优秀博士论文奖、2021年IEEE ICCC最佳论文奖和2015年IEEE WCSP最佳论文奖。他曾被评为2019年IEEE Communications Letters优秀编辑，并担任多个IEEE期刊的优秀审稿人。他现任IEEE Transactions on Communications、IEEE Communications Letters、IEEE Wireless Communications Letters的副编辑。他还担任IEEE Journal on Selected Areas in Communications首席客座编辑，是IEEE ICC 2019-2023和IEEE GLOBECOM 2020的研讨会联席主席，IEEE通信学会亚太地区青年协会主席。
Mohammad Reza Ghavidel Aghdam
博士
Özyeğin University
演讲标题:Toward Enhanced Wireless Communications: Leveraging Reconfigurable Intelligent Surfaces (RIS) and End-to-End Machine Learning
摘要: In the rapidly evolving landscape of wireless communication networks, the emergence of Reconfigurable Intelligent Surfaces (RIS) stands as a game-changer. These surfaces, empowered by intelligent algorithms, offer unprecedented control over electromagnetic wave propagation, promising to revolutionize communication efficiency and coverage. RIS introduces a paradigm shift in the way we conceive and implement wireless communication systems. By strategically deploying passive reflecting elements, RIS optimizes signal transmission, mitigates interference, and enhances overall network performance. However, to fully unlock the capabilities of RIS, seamless integration with machine learning is imperative.
End-to-end (E2E) machine learning algorithms provide a holistic approach to optimizing the performance of RIS-enabled communication networks. These algorithms leverage vast amounts of data to autonomously adapt RIS configurations, channel allocation, and beamforming strategies, thereby achieving optimal performance under dynamic and complex operating environments. Through E2E machine learning, RIS networks can continuously learn from real-world interactions, evolving to meet evolving communication demands.
In this presentation, we will explore the key challenges and opportunities in deploying RIS with E2E machine learning. We will discuss the intricacies of training machine learning models to effectively harness the capabilities of RIS, ensuring seamless integration with existing network infrastructures. Moreover, we will highlight promising applications of RIS-E2E machine learning in various domains, including 5G/6G networks, IoT connectivity, and smart city initiatives. Join us as we embark on a journey to unlock the full potential of RIS through E2E machine learning. Discover how this convergence of technologies is poised to redefine the future of wireless communication networks, ushering in an era of unprecedented connectivity, efficiency, and innovation.
简历: Dr. Mohammad Reza Ghavidel Aghdam 是一位杰出的学者和研究员，专门从事通信工程和无线通信领域。他于2014年从乌尔米亚大学获得电气工程学士学位，随后于2016年和2020年分别从伊朗的大不里士大学获得通信工程硕士和通信工程博士学位。从2017年开始，作为电气工程系的客座助理教授，在伊朗伊斯兰阿扎德大学伊尔希奇分校工程学院开始了他的学术之旅，Ghavidel Aghdam 博士迅速确立了自己作为一名敬业的教育者和研究员的地位。他对卓越的承诺和对推动该领域进步的热情，使他于2022年10月加入位于土耳其伊斯坦布尔的Özyeğin大学，成为备受推崇的通信理论与技术（CT&T）研究小组的博士后研究员。
Ghavidel Aghdam 博士的研究兴趣涵盖无线通信中一系列前沿话题，包括但不限于信号处理技术、智能反射表面（IRSs）或可重构智能表面（RISs）、机器学习应用、大规模多输入多输出（mMIMO）、毫米波（mmWave）通信以及非正交多址接入（NOMA）。他在这些领域的贡献显著推进了无线通信系统的理解和实施，为现代通信网络中的性能、效率和可扩展性的提升铺平了道路。
Yipeng Zhou
博士
Özyeğin University
演讲标题:Enhancing Federated Learning by Sparsifying Transmitted Model Updates
摘要:Federated learning facilitates the collaborative training of a machine learning model among geographically dispersed clients by exchanging model updates with a central server via Internet communication. However, transmitting these updates between the server and numerous decentralized clients over the Internet consumes considerable bandwidth and is susceptible to malicious attacks. This presentation showcases our various contributions aimed at improving communication efficiency and preserving privacy in federated learning. Our focus lies in sparsifying the transmission of model updates between the server and clients. By meticulously evaluating both the learning value, communication cost and privacy cost of transmitting each individual model update, we effectively mitigate the exposure of low-value updates to minimize communication and privacy costs. Extensive experiments conducted on real datasets demonstrate that our algorithms can significantly reduce communication costs and bolster privacy protection compared to the state-of-the-art federated learning baselines.
简历: Yipeng Zhou博士目前是澳大利亚麦考瑞大学科学与工程学院计算学院的高级讲师。在加入麦考瑞大学之前，他曾分别在南澳大学担任研究员和深圳大学担任讲师。他分别从香港中文大学获得博士和哲学硕士学位，以及从中国科学技术大学获得学士学位。他获得了2023年麦考瑞大学副校长研究卓越奖创新技术高度赞扬奖，以及2023年IEEE通信学会开放期刊最佳编辑奖。他还是2018年澳大利亚研究理事会发现早期职业研究者奖（DECRA）的获得者。他的研究兴趣包括联邦学习、数据隐私保护、网络等领域。他在顶级会议和期刊上发表了120多篇论文，包括IEEE INFOCOM、IJCAI、ICNP、IWQoS、IEEE ToN、JSAC、TPDS、TMC、TMM等
Dmitry A. Zaitsev
Professor
The University of Derby
演讲标题:Clans for HPC: Composition of clans to speed-up solving sparse linear systems on parallel and distributed architecture
摘要: Solving linear Diophantine systems of equations is applied in discrete-event systems, model checking, formal languages and automata, logic programming, cryptography, networking, signal processing, and chemistry. For modeling discrete systems with Petri nets, a solution in non-negative integer numbers is required, which represents an intractable problem. For this reason, solving such kinds of tasks with significant speedup is highly appreciated.
We introduce a nearness relation on a set of system’s equations, which transitive closure gives a clan relation. A sparse system is decomposed into a set of its clans. Solving a subsystem for each clan and then the clan composition system gives a speed-up of computations. We design a new solver of linear Diophantine systems, based on the simultaneous and parallel-sequential composition of the system clans, that runs on parallel architectures using a two level parallelization concept based on MPI and OpenMP. A decomposable system is usually represented by a sparse matrix; a minimal clan size of the decomposition restricts the granulation of the technique. A dynamic task-dispatching subsystem is developed for distributing systems on nodes in the process of compositional solution. Computational speedups are obtained on a series of test examples, e.g., illustrating that the best value constitutes up to 45 times speedup obtained on 5 nodes with 20 cores each. For load balancing, aggregation of the minimal clans has been implemented, that yields an additional speed-up. Solving sparse systems over fields of real numbers, using SVD decomposition to obtain basis solutions for clans, also reveals considerable speed-up on real-life tasks from the MatrixMarket repository.
Basic references:
Dmitry A. Zaitsev, Tatiana R. Shmeleva, Piotr Luszczek. Aggregation of clans to speed-up solving linear systems on parallel architectures, International Journal of Parallel, Emergent and Distributed Systems, 37(2), 2022, 198-219.http://dx.doi.org/10.1080/17445760.2021.2004412
Dmitry Zaitsev, Stanimire Tomov, Jack Dongarra. Solving Linear Diophantine Systems on Parallel Architectures, IEEE Transactions on Parallel and Distributed Systems, 30(5), 2019, 1158-1169. http://dx.doi.org/10.1109/TPDS.2018.2873354
Zaitsev D.A. Sequential composition of linear systems' clans, Information Sciences, 363, 2016, 292-307. http://dx.doi.org/10.1016/j.ins.2016.02.016
简历: Dmitry A. Zaitsev（英国德比大学）开发了具有规则结构的无限Petri网的分析、Petri网的部落分解、元胞自动机的广义邻域以及基于表格的模糊逻辑函数合成方法。他设计了用于制造操作计划和控制的Opera-Topaz软件；一种新的网络协议栈E6，并在Linux内核中实现了该协议；以及Petri网分析软件Deborah、Adriana和ParAd。他还开发了TCP、BGP、IOTP协议以及Ethernet、IP、MPLS、PBB和蓝牙网络的模型。他目前的研究兴趣包括Petri网理论及其在网络、计算和自动化制造中的应用。最近，他开始在百亿次计算领域工作，应用其部落理论来加速在并行和分布式架构上求解稀疏线性系统，并开发了一种新的Sleptsov网计算范式。他曾是与中国和奥地利联合项目的共同主任。最近，他作为访问教授在德国多特蒙德工业大学获得DAAD奖学金，美国田纳西大学诺克斯维尔分校获得富布赖特奖学金，荷兰埃因霍温理工大学，奥地利约翰内斯·开普勒大学，法国尼斯蔻特·达祖尔大学，以及德国达姆施塔特工业大学任职。他出版了一部专著、四个书章节以及一百多篇论文，包括JCR和CORE A期刊的文章。他是ACM和IEEE的高级会员。
Jiashen Teh
Associate Professor
Universiti Sains Malaysia (USM)
演讲标题:Dynamic Line Rating (DLR) for Enhanced Grid Reliability
摘要: In this keynote speech, dynamic thermal line rating (DTLR) emerges as a transformative solution in modernizing power grid management. Highlighting its significance in enhancing grid resilience and efficiency, the speaker delves into the principles and applications of DTLR technology. Through real-world case studies and innovative approaches, attendees gain insights into how DTLR dynamically optimizes transmission line capacities based on environmental conditions, mitigating congestion risks and enabling higher utilization rates. Furthermore, the speech explores the integration of DTLR into smart grid frameworks, paving the way for a more adaptive and sustainable energy infrastructure.
简历: Teh 博士于2016年获得英国曼彻斯特大学电气工程博士学位，目前是马来西亚理科大学（USM）的副教授。他也是台湾UPE-Power的技术总监。他主要研究灵活的输电线路额定功率在提高电网可靠性方面的优势。截至目前，他在SCIE数据库中发表了60多篇期刊文章，其中大多数位列排名的前两个四分位，在Google Scholar上获得了超过2900次引用，h指数为31。他在2019年、2020年和2021年被斯坦福大学评为全球各领域被引用次数最多的前2%研究人员之一。在2021年和2022年，他分别获得了IEEE电力与能源马来西亚分会的杰出工程师奖和IET马来西亚分会的杰出青年专业人士奖。