自动化学报2018，Vol.44Issue(2)：193-215,23.DOI:10.16383/j.aas.2018.c160812
实值多变量维数约简:综述
Real-valued Multivariate Dimension Reduction: A Survey
摘要
Abstract
As one of the classical problems in machine learning,dimension reduction is used for dealing with the curse of dimensionality,speeding up computational efficiency of the algorithm,and improving interpretability as well as visualizing high-dimensional data.Traditional dimension reduction algorithms such as principal component analysis (PCA) and linear discriminant analysis are mainly suitable for unlabeled data or classification data.When the response variables are univariate or multivariate continuous real-valued ones,however,such dimension reduction methods cannot guarantee the effective predictive performance of the reduced subspace.In the recent two decades,researchers have been devoted to studying this issue with different viewpoints,attaining many promising and systemic achievements.Under this background,we will survey the developments of real-valued multivariate dimension reduction in detail.We will also introduce its basic concepts,algorithms and theories,and discuss some potential research directions deserving investigating.关键词
维数约简/维数灾难/回归分析/条件独立性/互信息Key words
Dimension reduction/curse of dimensionality/regression analysis/conditional independence/mutual information引用本文复制引用
单洪明,张军平..实值多变量维数约简:综述[J].自动化学报,2018,44(2):193-215,23.基金项目
国家自然科学基金(61673118),上海市浦江人才计划(16PJD009)资助 Supported by National Natural Science Foundation of China (61673118) and Shanghai Pujiang Program (16PJD009) （61673118）
