与Bernd Heidergott教授面对面交流

6月8日下午主题：

The Measure-valued Differentiation Approach to Gradient Estimation

摘要：Measure-valued differentation (MVD) is next to IPA and the Score

Fuction method one of the three main techniques for estatblishing unbiased

gradient estimators. We give a gentle introduction into MVD and illustrate

its relation with IPA and the Score Function.

主讲人简介：Bernd Heidergott is Chair of Stochastic Optimization at the Department of Econometrics and Operations Research at the Faculty of Economics and Business Administration of the Vrije Universiteit Amsterdam. At this university he is program director of the bachelor and master program Econometrics and Operations Research, and member of the board of the Amsterdam Business Research Institute (ABRI). Before joining the VU University Amsterdam he held post-doc positions at Erasmus University Rotterdam, Eindhoven University of Technology, Delft University of Technology, and EURANDOM. He is author of two monographs and more than 40 international journal papers. Bernd Heidergott is ABRI Research Fellow, and Tinbergen Institute Fellow.