当前位置: 首页 / English / Academics / 正文

Adaptive multi-fidelity surrogate modeling for Bayesian inference in inverse problems

作者:   时间:2020-12-14   点击数:

Keynote Speaker:Zhou Tao

Abstract:

The generalized polynomial chaos (gPC) are widely used as surrogate models in Bayesian inference to speed up the Markov chain Monte Carlo simulations. However, the use of gPC-surrogates introduces model errors that may severely distort the estimate of the posterior distribution. In this talk, we present an adaptive procedure to construct an adaptive gPC-surrogate. The key idea is to refine the surrogate over a sequence of samples adaptively so that the surrogate is much more accurate in the posterior region. We then introduce an adaptive surrogate modeling approach based on deep neural networks to handle problems with high dimensional parameters.

Speaker Introduction:

Zhou Tao is an Associate Researcher of Academy of Mathematics and Systems Science of Chinese Academy of Sciences. He has been engaged in post-doctoral research in the Swiss Federal Institute of Technology in Lausanne. The main research directions are uncertainty quantification, stochastic optimal control and parallel-in-time algorithm. He has published more than 50 papers in international top journals such asSIAM ReviewandSINUM. In 2016, he was awarded the China Youth Science and Technology Award of China Society for Industrial and Applied Mathematics, and in 2018, he was granted "Excellent Young Scientists Fund" from the National Natural Science Foundation of China. He has been the Associate Editor in Chief ofInternational Journal for UQsince 2017 as well as the Associate Editor of several international journals such asCommun. Comput. Phys.andNumerical Mathematics:TMA, etc.. Since 2020, he has been the Associate Editor ofSIAM Journal on Scientific Computingwhich is a top journal of computational mathematics. Since 2018, he has been the Chief Scientist of "uncertainty assessment method of complex system model" in the subject field of scientific challenges of COSTIND.

Inviter:

Zhao Weidong ProfessorofSchool of Mathematics

Time:

9:00-10:00 on December 22 (Tuesday)

Location:

Tencent Meeting, ID: 585 798 369

https://meeting.tencent.com/s/HMGBsrd6HrFT

Hosted by: School of Mathematics, Shandong University

地址:中国山东省济南市山大南路27号   邮编:250100  

电话:0531-88364652  院长信箱:sxyuanzhang@sdu.edu.cn

Copyright@山东大学数学学院

微信公众号