Unlock instant, AI-driven research and patent intelligence for your innovation.

A quantitative evaluation method of matrix sparsity

A quantitative evaluation and sparsity technology, applied in the field of quantitative evaluation of matrix sparsity, can solve the problem of unobjective evaluation of sparsity, and achieve the effect of objective and accurate evaluation

Inactive Publication Date: 2019-02-26
XI AN JIAOTONG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the actual sparse model, it is hoped that the difference between the local part and the whole should be as obvious as possible, that is, the difference between the elements of the matrix with the same non-zero number should be as large as possible, so for the sparse model, according to custom indicator sparseness(x j ) for sparsity evaluation is not objective

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A quantitative evaluation method of matrix sparsity
  • A quantitative evaluation method of matrix sparsity
  • A quantitative evaluation method of matrix sparsity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The present invention is described in detail below in conjunction with accompanying drawing,

[0024] refer to figure 1 , a quantitative evaluation method for matrix sparsity, comprising the following steps:

[0025] 1) Obtain the matrix X to be evaluated in the sparse model m×n , m represents the number of matrix rows, n represents the number of matrix columns;

[0026] 2) Calculate the matrix X to be evaluated m×n The mean of each column mean(x j ), variance var(x j ) and the maximum value max(x j );

[0027] Indicates the matrix X to be evaluated m×n Mean of column j, where i=1, 2, ..., m, j = 1, 2, ..., n, X i，j is the element in row i and column j of the matrix; var(x j ) = Indicates the matrix X to be evaluated m×n The sample variance of column j; max(x j ) represents the matrix X to be evaluated m×n the maximum value of column j;

[0028] 3) Determine the matrix X to be evaluated m×n The number of sparse elements in each column z j ;

[0029] ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A quantitative evaluation method for matrix sparsity is disclosed. Firstly, the matrix to be evaluated in the sparse model is obtained. And then calculating the mean value, variance and maximum valueof each column of the matrix to be evaluated; Determining the number of sparse elements in each column of the matrix to be evaluated; Then the relative sparsity of each column of the matrix to be evaluated is calculated. Finally, the global sparsity of the matrix to be evaluated is calculated. The method of the invention makes up for the defects that the existing index evaluation is not comprehensive enough and the local difference is ignored, and considers the influence of the mean variance on the sparsity of the matrix and the contribution of the non-zero smaller value of the matrix to the sparsity, so that the evaluation of the sparsity of the matrix is more objective and accurate.

Description

technical field [0001] The invention belongs to the technical field of sparse models, and in particular relates to a quantitative evaluation method for matrix sparsity. Background technique [0002] In recent years, the sparse model has been greatly developed and applied in the fields of signal processing, image processing, pattern recognition, and target detection. The sparse model removes a large number of redundant variables and only retains the most relevant explanatory variables to the response variable, simplifying the While retaining the most important information in the data set, it effectively solves many problems in high-dimensional data set modeling. At the same time, the sparse model has better interpretability, which is convenient for data visualization, reducing the amount of calculation and transmission and storage. [0003] The performance evaluation of the sparse model usually lies in the sparsity of the decomposed matrix and whether it is faithful to the o...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/46G06F17/16
CPCG06F17/16G06V10/40G06V10/513
Inventor 梁霖山磊刘飞陈元明栗茂林徐光华
Owner XI AN JIAOTONG UNIV