報告題目：Optimal design criteria for computer experiments
報 告 人：孫法省教授 東北師範大學
報告時間：2020年7月9日 9:00-10:00
報告地點：騰訊會議 ID：380 126 085
點擊鍊接入會，或添加至會議列表：
https://meeting.tencent.com/s/51Ymc8FgbuU0
校内聯系人：朱複康 fzhu@jlu.edu.cn
報告摘要：
Efficient designs are in high demand in practice for both computer and physical experiments. Space-filling designs are widely used in both computer and physical experiments. Existing designs may have bad low-dimensional projections, which is undesirable when only a few factors are active. We propose a new design criterion, called uniform projection criterion, by focusing on projection uniformity. We show that the average squared correlation metric is a function of the pairwise L2-distances between the rows only. We further explore the connections among column-orthogonality, maximin distance and projection uniformity. Based on these connections, we develop new lower and upper bounds for column-orthogonality and projection uniformity from the perspective of distance between design points. These results not only provide new theoretical justifications for each criterion but also help in finding better space-filling designs under multiple criteria. An application of uniform projection designs via a multidrug combination experiment also be given.
報告人簡介：
孫法省，東北師範大學教授、博導，吉林省優秀教師。主要從事大數據抽樣與分析、計算機試驗設計與分析、及高維數據分析等方面的研究。獲教育部自然科學二等獎、全國統計科學研究優秀成果獎、吉林省自然科學學術成果獎各一項。
