报告题目：Multiview PCA: A methodology of feature extraction and dimension reduction for high-order data

报告人：夏志明 教授 陕西师范大学

照片：

邀请人：薄立军

报告时间：2023年4月13日上午10:00-11:30

报告地点：南校区网安大楼会议中心121会议室

报告人简介：统计学博士，教授，博士生导师，西北大学数学学院副经理，西北大学现代统计研究中心副主任，陕西省统计协会常务理事，主要致力于张量数据分析、大数据异质性结构推断、分布式统计推断与计算、生物统计学等数据科学理论与应用研究。在“Biometrika”、“Journal of machine learning research”,“Technometrics”、“IEEE Transaction of Cybernetics”、“Statistics in Medicine”等国际统计与机器学习期刊以及“中国科学”、“应用概率统计”等国内期刊发表论文40余篇；主持国家自然科学基金项目4项，主持省部级项目3项,作为骨干成员获得“陕西省科学技术进步奖”二、三等奖共2项，“陕西省高校科学技术奖”一等奖共2项，“陕西省国防科技进步奖”一等奖1项；先后赴香港科技大学、佛罗里达大学等科研机构进行专业访问与学术交流。

报告摘要：In this talk, we focus on a new PCA methodology for tensor data. Facing with rapidly-increasing demands for analyzing high-order data or multiway data, feature-extracting methods become imperative for analysis and processing. We propose a more flexible and more powerful tool, called the Multiview Principal Components Analysis (Multiview-PCA) . By segmenting a random tensor into equal-sized subarrays namedsectionsand maximizing variations caused by orthogonal projections of thesesections, the Multiview-PCA finds principal components in a parsimonious and flexible way. In so doing, two new operations on tensors, thedirection inner/outer product, are introduced to formulate tensor projection and recovery. We also propose an adaptive depth and direction selection algorithm for implementation of Multiview-PCA. The Multiview-PCA is then tested in terms of subspace recovery ability, compression ability and feature extraction performance when applied to a set of artificial data, surveillance videos and hyperspectral imaging data.