The Past,Present,and Future of Monte Carlo Methods for Statistical Signal Processing（2）

2014 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC 2014)——In the past two and a half decades,Monte Carlo-based methods have become one of the most important approaches for resolving highly challenging problems in statistical signal processing.They are computationally intensive and are based on repeated generation of random samples.To the generated samples,one assigns weights so that one can approximate various probability distributions of interest.The estimated distributions are subsequently used for statistical inference.The Monte Carlo-based methods are widely used in both batch-type and dynamic settings.In this presentation,the aim is to provide an overview of these methods,their history and their state-of-the art.The aim,too,is to discuss future directions of research with these methods.

关键词： 蒙特卡罗方法 统计模拟 信号处理 The 2014 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC 2014)

主讲人：IEEE FELLOW Petar M.Djurić 机构：Department of Electrical and Computer Engineering, State University of New York at Stony Brook

时长：0:20:41 年代：2014年