4月26日（周一）上午9：30，美国北卡罗莱纳大学戴怀宇教授将在信电楼215进行学术报告，欢迎各位老师和同学们参加。

 

Title：Distributed Inference in Networked Systems through Structured Variational Methods

 

Speaker：     

Prof. Huaiyu Dai (戴怀宇)        

Associate Professor

Department of Electrical and Computer Engineering,

North Carolina State University, USA

Abstract：

Reasoning and learning in complex networked systems are often casted as probabilistic inference in a graphical model. In such systems, finding solutions in a distributed fashion is often of great interest and importance. In this talk, we will introduce our recent work on exploring structured variational methods towards developing a general and flexible framework for distributed inference in networked systems, simultaneously exploiting the simplicity of variational methods such as the mean-field approach (for inter-cluster processing) and the accuracy of more complex inference methods such as the belief-propagation algorithm (for intra-cluster processing). We will also discuss associated performance analysis, which involves a mixed Markov process on both vertices and directed edges. Finally we will address a relevant research line on designing distributed clustering schemes that better match the needs of inference tasks.

 

Bio:

Huaiyu Dai received the B.E. and M.S. degrees in electrical engineering from Tsinghua University, Beijing, China, in 1996 and 1998, respectively, and the Ph.D. degree in electrical engineering from Princeton University, Princeton, NJ, USA in 2002. Currently he is an Associate Professor of Electrical and Computer Engineering at NC State University, Raleigh, NC, USA. His research interests are in the general areas of communication systems and networks, advanced signal processing for digital communications, and communication theory and information theory.

He has served as an associate editor of IEEE Transactions on Wireless Communications, and is currently an associate editor of IEEE Transactions on Signal Processing. He has co-edited two special issues for EURASIP journals on distributed signal processing techniques for wireless sensor networks, and on multiuser information theory and related applications, respectively.