Conditional Central Limit Theorems for Gaussian Projections

被引:0
作者
Reeves, Galen [1 ,2 ]
机构
[1] Duke Univ, Dept ECE, Durham, NC 27706 USA
[2] Duke Univ, Dept Stat Sci, Durham, NC 27706 USA
来源
2017 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT) | 2017年
关键词
LOW-DIMENSIONAL PROJECTIONS; DISTRIBUTIONS;
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper addresses the question of when projections of a high-dimensional random vector are approximately Gaussian. This problem has been studied previously in the context of high-dimensional data analysis, where the focus is on low-dimensional projections of high-dimensional point clouds. The focus of this paper is on the typical behavior when the projections are generated by an i.i.d. Gaussian projection matrix. The main results are bounds on the deviation between the conditional distribution of the projections and a Gaussian approximation, where the conditioning is on the projection matrix. The bounds are given in terms of the quadratic Wasserstein distance and relative entropy and are stated explicitly as a function of the number of projections and certain key properties of the random vector. The proof uses Talagrand's transportation inequality and a general integral-moment inequality for mutual information. Applications to random linear estimation and compressed sensing are discussed.
引用
收藏
页数:5
相关论文
共 25 条
[1]  
[Anonymous], 2014, Estimation in high dimensions: a geometric perspective
[2]   The central limit problem for convex bodies [J].
Anttila, M ;
Ball, K ;
Perissinaki, I .
TRANSACTIONS OF THE AMERICAN MATHEMATICAL SOCIETY, 2003, 355 (12) :4723-4735
[3]  
Bobkov SG, 2003, ANN PROBAB, V31, P195
[4]  
Dasgupta S., 2006, C UNC ART INT
[5]   ASYMPTOTICS OF GRAPHICAL PROJECTION PURSUIT [J].
DIACONIS, P ;
FREEDMAN, D .
ANNALS OF STATISTICS, 1984, 12 (03) :793-815
[6]  
DIACONIS P, 1987, ANN I H POINCARE-PR, V23, P397
[7]   Message-passing algorithms for compressed sensing [J].
Donoho, David L. ;
Maleki, Arian ;
Montanari, Andrea .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2009, 106 (45) :18914-18919
[8]  
Dumbgen Lutz, 2013, From Probability to Statistics and Back: High-Dimensional Models and Processes-A Festschrift in Honor of Jon A. Wellner, P91
[9]  
Eldar Y.C., 2012, Compressed Sensing: Theory and Applications
[10]   ON ALMOST LINEARITY OF LOW-DIMENSIONAL PROJECTIONS FROM HIGH-DIMENSIONAL DATA [J].
HALL, P ;
LI, KC .
ANNALS OF STATISTICS, 1993, 21 (02) :867-889