马氏过程的集中不等式

Release time:2023-04-06Views:305

Title马氏过程的集中不等式


Speaker: Liming WuHarbin Institute of Technologyl'Université Clermont-Auvergne


Time: Tuesday, April 11, 2023, 16:00-17:00


Location: Mingde Building B201

Zoom ID: 832 6498 6834Password: 2023


Abstract: 这个报告的主题是马氏过程观察量的经验均值趋于真实均值的集中不等式,或者说随机算法的错误概率的绝对估计。我将介绍与集中不等式等价的“运费vs Fisher信息量”不等式,以及怎样使用PoincarelogSobolev不等式,与最优传输,熵这些工具去证明“运费vs Fisher信息量”不等式。作为应用例子,我将介绍高维分布Gibbs算法的错误概率绝对估计。


 

More information about Colloquium Talks can be found here.

 


Copyright (C)2017 Institute for Advanced Study in Mathematics of HIT All Rights Reserved.
Recruitment:
Contact Us:
Tel:86413107      Email:IASM@hit.edu.cn
Add:NO.92 West Da Zhi St. Harbin China
Technical support:Net & Information Center,HIT