On graphical models and convex geometry

被引：2

作者：

Bar, Haim ^{[1
]}

Wells, Martin T. ^{[2
]}

机构：

[1] Univ Connecticut, Dept Stat, Room 315,Philip E Austin Bldg, Storrs, CT 06269 USA

[2] Cornell Univ, Dept Stat & Data Sci, 1190 Comstock Hall, Ithaca, NY 14853 USA

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2023年 / 187卷

关键词：

Convex geometry; Correlation matrix estimation; Expectation Maximization (EM) algorithm; Graphical models; Grassmann manifold; High-dimensional inference; Network models; Phase transition; Quasi-orthogonality; Two-group model; COVARIANCE MATRICES; MAXIMUM-LIKELIHOOD; CONFIDENCE-REGIONS; SPARSE REGRESSION; SELECTION; DISTRIBUTIONS; REGULARIZATION; COHERENCE; INFERENCE; RECOVERY;

D O I：

10.1016/j.csda.2023.107800

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

A mixture-model of beta distributions framework is introduced to identify significant correlations among P features when P is large. The method relies on theorems in convex geometry, which are used to show how to control the error rate of edge detection in graphical models. The proposed 'betaMix' method does not require any assumptions about the network structure, nor does it assume that the network is sparse. The results hold for a wide class of data-generating distributions that include light-tailed and heavy-tailed spherically symmetric distributions. The results are robust for sufficiently large sample sizes and hold for non-elliptically-symmetric distributions. (c) 2023 Elsevier B.V. All rights reserved.

引用

页数：17

共 50 条

[1] Algorithms for Convex Hull Finding in Undirected Graphical Models
Heng, Pei
Sun, Yi
APPLIED MATHEMATICS AND COMPUTATION, 2023, 445
[2] GEOMETRY OF MAXIMUM LIKELIHOOD ESTIMATION IN GAUSSIAN GRAPHICAL MODELS
Uhler, Caroline
ANNALS OF STATISTICS, 2012, 40 (01): : 238 - 261
[3] LOCALLY ASSOCIATED GRAPHICAL MODELS AND MIXED CONVEX EXPONENTIAL FAMILIES
Lauritzen, Steffen
Zwiernik, Piotr
ANNALS OF STATISTICS, 2022, 50 (05): : 3009 - 3038
[4] Efficient Low Rank Convex Bounds for Pairwise Discrete Graphical Models
Durante, Valentin
Katsirelos, George
Schiex, Thomas
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162, 2022,
[5] Edge selection based on the geometry of dually flat spaces for Gaussian graphical models
Hirose, Yoshihiro
Komaki, Fumiyasu
STATISTICS AND COMPUTING, 2013, 23 (06) : 793 - 800
[6] Edge selection based on the geometry of dually flat spaces for Gaussian graphical models
Yoshihiro Hirose
Fumiyasu Komaki
Statistics and Computing, 2013, 23 : 793 - 800
[7] Sparse graphical models via calibrated concave convex procedure with application to fMRI data
Son, Sungtaek
Park, Cheolwoo
Jeon, Yongho
JOURNAL OF APPLIED STATISTICS, 2020, 47 (06) : 997 - 1016
[8] Transforming Graphical System Models to Graphical Attack Models
Ivanova, Marieta Georgieva
Probst, Christian W.
Hansen, Rene Rydhof
Kammueller, Florian
GRAPHICAL MODELS FOR SECURITY, GRAMSEC 2015, 2016, 9390 : 82 - 96
[9] Geometry of Convex Geometries
Chalopin, Jeremie
Chepoi, Victor
Knauer, Kolja
DISCRETE & COMPUTATIONAL GEOMETRY, 2025,
[10] Convex and Discrete Geometry
Cobzas, S.
STUDIA UNIVERSITATIS BABES-BOLYAI MATHEMATICA, 2007, 52 (04): : 164 - 165

← 1 2 3 4 5 →