报告人:郭庆华
报告题目:Sparse Bayesian Learning Using Approximate Message Passing with Unitary Transformation
报告时间:11月8日 10:00
报告地点:4D322
报告简介:
Sparse Bayesian learning (SBL) has emerged as an attractive method for compressive sensing or sparse signal recovery. The conventional SBL algorithm suffers from high computational complexity in large scale problems. Recently, SBL has been implemented with low complexity based on the approximate message passing (AMP) algorithm. However, it is vulnerable to ‘difficult’ measurement matrices as AMP can easily diverge. Damped AMP has been used to alleviate the problem at the cost of significantly slowing the convergence rate. In this talk, I will introduce a new low complexity SBL algorithm, which is designed based on the AMP with unitary transformation (UTAMP). I will show that, compared to state-of-the-art AMP based SBL algorithms, our proposed UTAMP-SBL is much more robust and converges much faster, leading to remarkably better performance. In many cases, the performance of the algorithm can approach the support-Oracle MMSE bound closely.
报告人简介:
郭庆华,博士,澳大利亚首届优秀青年基金获得者,于2008年11月获香港城市大学博士学位,现任职于澳大利亚伍伦贡大学(University of Wollongong)电气、计算机和通信工程学院副教授和西澳大利亚大学工程学院兼职副教授。郭庆华的研究领域包括控制科学与工程、信号处理,通信和机器学习。