Sparse subset selection method based on dissimilarity and Laplace regularization

A technique of subset selection and similarity, which is applied in instruments, character and pattern recognition, computer components, etc.

Inactive Publication Date: 2020-05-01
杨晨曦
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Second, although some practical datasets do not exist in vector spaces, such as social network data or proteomics data, pairwise relationships can already be efficiently computed

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
  • Sparse subset selection method based on dissimilarity and Laplace regularization
  • Sparse subset selection method based on dissimilarity and Laplace regularization
  • Sparse subset selection method based on dissimilarity and Laplace regularization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] Suppose we have a source set X = {x 1 ,...,x M} and a target set Y={y 1 ,...,y N}, they contain M and N elements respectively, assuming we get the dissimilarity relationship between X and Y d ij means x i stands for y j The degree of good or bad, the smaller its value means x i The better it can represent y j . Write this binary relationship in matrix form as follows

[0032]

[0033] Our goal is to find a smaller subset of X such that it can well represent the target set Y, such as figure 1 shown, where figure 1 Left: the dissimilarity relationship between the source set X and the target set Y; right: a subset of the source set X is found, which can well represent the characteristics of the target set Y.

[0034] Given a dissimilarity matrix D, we need to find a representative subset of the source set X, that is, the representative unit, so that it can effectively represent the target set Y. To this end, we consider the dissimilarity with d ij The asso...

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

The invention discloses a sparse subset selection method based on dissimilarity and Laplace regularization. According to the method, the problem that representative elements capable of effectively representing a target set are found from a source set is considered by utilizing a pairwise dissimilarity relationship between a given source set and the target set, and a dissimilarity-based low-rank sparse subset selection model is proposed, so that the problem can be effectively solved by using convex programming. On the basis of previous work, the structure between representative elements is considered, so that the number of the representative elements is smaller, and the representative quality is higher. Algythm 1 is also used for effectively realizing the algorithm, and the algorithm can befurther parallelized, so that the calculation time can be further reduced.

Description

technical field [0001] This application relates to the field of machine learning and data analysis, in particular to a sparse subset selection method based on dissimilarity and Laplacian regularization. Background technique [0002] Sparse subset selection: Discovering a large number of models or subsets of data points that preserve the characteristics of the entire set is an important problem in machine learning and data analysis in computer vision applications, such as image and natural language processing, biological There are a large number of applications in health informatics, recommendation systems, etc. These information elements are called representatives or demonstrations. Data representation helps summarize and visualize datasets of text / web documents, images and videos, thus increasing the interpretability of large-scale datasets for data analysts and domain experts. Model representations help to efficiently describe complex phenomena or events using a small nu...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62
CPCG06F18/24
Inventor 杨晨曦
Owner 杨晨曦
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap