Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

LS-SVMs online learning method based on budget support vector set

A technology that supports vector sets and learning methods, applied in the fields of data mining and machine learning, can solve the problems of reducing model storage space and inability to effectively control model size, etc.

Inactive Publication Date: 2016-07-20
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a kind of LS-SVMs online learning method based on the budget support vector set, which can reduce the storage space of the model and reduce the running time. time to meet the real-time requirements of application problems

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
  • LS-SVMs online learning method based on budget support vector set
  • LS-SVMs online learning method based on budget support vector set
  • LS-SVMs online learning method based on budget support vector set

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] Embodiment 1: Take the double helix classification problem as an example for illustration. Such as figure 1 Shown is a schematic diagram of a LS-SVMs online learning method based on a budget support vector set provided according to an embodiment of the present invention, and the online learning method contains the following steps:

[0054] Step 1: Determine the budget range using training samples. The specific steps are:

[0055] (1) Select the data set to be processed. In this embodiment, the reference data set Adult9 is taken as an example for illustration. Take the sample block size as 1, randomly select 2000 sample points from Adult9 to construct the training set and select 1000 sample points to construct the test set, and apply the standardized formula Preprocess the sample points. Choose Gaussian Radial Basis Functions As a kernel function, the kernel width parameter σ takes the default value, which is the dimension of the sample point.

[0056] (2) Determ...

Embodiment 2

[0079] Embodiment 2: Take the application of the sliding window technology as an example. The difference from Embodiment 1 is that in this embodiment, in step 3, the data stream is collected in the form of one-by-one, and the predictor is used to predict the samples in the data stream; in step 4, when collecting After getting the real label of the sample, compare the prediction output of the predictor, ignore the correct prediction sample, record the wrong prediction sample, add the wrong prediction sample to the support vector set, and remove the earliest support vector in the support vector set according to the time criterion to maintain The budget is stable.

[0080] Figures 5 to 8 In order to adopt the online learning method of the present invention and the existing stochastic budget perceptron method and the online gradient descent method to process the average online Schematic diagram of test accuracy comparison, by Figures 5 to 8 It can be seen that the test accura...

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 relates to an LS-SVMs online learning method based on a budget support vector set. A budge scope is determined on a training set, an initial support vector set is selected, an LS-SVMs model is established, the LS-SVMs model is solved by use of a conjugate gradient method, an online predicitor is obtained by updating the LS-SVMs model by use of a low-rank matrix correction method and a Sherman-Morrison-Woodbury formula, and thus online prediction of a data flow is realized. The method employs a fixed budge strategy, can effectively control the scale of an online learning model, saves storage space, and is low in calculation complexity and easy to realize. The online learning method provided by the invention has the following advantages: the problem of online application with a data flow feature can be flexibly processed, data can be collected in the form of data blocks, compared to a conventional batch processing mode and a conventional online learning mode, the calculation complexity and the model operation time are substantially reduced, the regression problem and the classification problem can be simultaneously resolved, and the selection problem of the LS-SVMs model can be highly efficiently solved.

Description

technical field [0001] The invention belongs to the field of data mining and machine learning, and relates to data mining and data processing methods, in particular to an LS-SVMs online learning method based on budget support vector sets. Background technique [0002] LS-SVMs are typical kernel learning models built on the principle of structural risk minimization. As a convex optimization problem with a good smoothing structure, the KKT condition of LS-SVMs model can be transformed into a special saddle point system with (2,2) blocks of order 1. The current mainstream methods for solving the LS-SVMs model mainly include: (1) use the preprocessor to convert the above saddle point system into two positive definite systems equivalently, and use the classic conjugate gradient algorithm to solve the positive definite system; (2) transform the saddle point system Expressed as two linear equations, and using the null space method to solve; (3) using the minimum residual method to...

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): G06K9/62
CPCG06F18/2411
Inventor 渐令宋允全申淑谦梁锡军
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com