Dimension Estimation Using Random Connection Models

被引:0
作者
Serra, Paulo [1 ,2 ]
Mandjes, Michel [2 ]
机构
[1] Eindhoven Univ Technol, Dept Math & Comp Sci, Groene Loper 5,MetaForum Bldg, NL-5612 AZ Eindhoven, Netherlands
[2] Univ Amsterdam, Kortewag de Vries Inst Math, Sci Pk 105-107, NL-1098 XG Amsterdam, Netherlands
关键词
adaptation; dimensionality reduction; intrinsic dimension; random connection model; random graph; INTRINSIC DIMENSIONALITY; PROXIMITIES;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Information about intrinsic dimension is crucial to perform dimensionality reduction, compress information, design efficient algorithms, and do statistical adaptation. In this paper we propose an estimator for the intrinsic dimension of a data set. The estimator is based on binary neighbourhood information about the observations in the form of two adjacency matrices, and does not require any explicit distance information. The underlying graph is modelled according to a subset of a specific random connection model, sometimes referred to as the Poisson blob model. Computationally the estimator scales like n log n, and we specify its asymptotic distribution and rate of convergence. A simulation study on both real and simulated data shows that our approach compares favourably with some competing methods from the literature, including approaches that rely on distance information.
引用
收藏
页数:35
相关论文
共 54 条
[1]  
Abrahao B, 2008, IMC'08: PROCEEDINGS OF THE 2008 ACM SIGCOMM INTERNET MEASUREMENT CONFERENCE, P157
[2]   NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].
AKAIKE, H .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723
[3]  
[Anonymous], 2004, Advances in neural information processing systems, DOI DOI 10.5555/2976040.2976138
[4]  
[Anonymous], 2015, KDD
[5]  
[Anonymous], 1982, PATTERN RECOGN
[6]  
[Anonymous], P 1 WORKSH GEN SIGN
[7]  
[Anonymous], P GEN SIGN PROC STAT
[8]   INTRINSIC DIMENSIONALITY OF SIGNAL COLLECTIONS [J].
BENNETT, RS .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1969, 15 (05) :517-+
[9]  
Beyer K, 1999, LECT NOTES COMPUT SC, V1540, P217
[10]   Some theory for Fisher's linear discriminant function, 'naive Bayes', and some alternatives when there are many more variables than observations [J].
Bickel, PJ ;
Levina, E .
BERNOULLI, 2004, 10 (06) :989-1010