报告题目：More than Prediction: the role of Statistics in Industrial R&D in the Physical Sciences 

报告人：Mark Wolters

报告时间：2021年3月25日 14:00-15:00

报告地点：上海数学中心102室（江湾）

报告摘要：

In industries dependent on the physical sciences, it is common to encounter situations where i) data may be scarce or expensive, ii) the cost of incorrect predictions may be high, or iii) a study objective is to gain mechanistic understanding.  The black-box predictive modelling techniques of machine learning are not sufficient for handling such cases--statistical modelling is a more natural and beneficial approach.  At the same time, developments in machine learning, computer simulation, and robotics have created many new possibilities for integration of statistical techniques in the broader context of "digitalization of industry." For this reason the field of industrial statistics is currently very diverse and interesting.  This talk will be a high-level overview of some topics and trends in this area, with particular focus on areas where statisticians can make a contribution.

报告人简介：

After receiving his bachelor's degree in Metals and Materials Engineering from the University of British Columbia, Mark Wolters had a first career as a professional engineer.  He later chose to pursue graduate studies in Statistics, culminating in a Ph.D. and postdoctoral fellowship at the University of Western Ontario. He worked as a Young Investigator at Fudan University's Shanghai Center for Mathematical Sciences from 2014 until 2019.  His statistical research has related to model selection, probability density estimation, and autologistic regression.  Since 2019 he has worked on a diverse range of data science projects as a digitalization and modelling scientist at BASF Advanced Chemical Corporation.