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

Rapid neural network leaning method

A neural network learning and extremely fast learning machine technology, applied in the field of artificial intelligence, can solve problems such as overfitting, performance impact, ELM does not consider the weight of the error, etc.

Inactive Publication Date: 2013-04-24
XIAN UNIV OF POSTS & TELECOMM
View PDF1 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] But ELM is based on the principle of empirical risk minimization, which may lead to overfitting problem [6]
In addition, because ELM does not consider the weight of the error, its performance will be seriously affected when there are outliers in the data set [7]

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
  • Rapid neural network leaning method
  • Rapid neural network leaning method
  • Rapid neural network leaning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0102] Here we compare the performance of RELM, ELM, BP and Support Vector Machine (Support Vector Machine, SVM) [13, 14] through experiments. The execution environment of RELM, ELM and BP is Matlab7.0, and the execution environment of SVM is C language. RELM is implemented by ourselves. The source code of ELM can be directly downloaded from Huang's personal homepage1, and the BP algorithm has been integrated in the neural network toolbox that comes with Matlab and can be used directly. There are many variants of the BP algorithm, and we choose the fastest Levenberg-Marquardt algorithm for experimentation. SVM algorithm We use the SVM package implemented in C language: LibSVM2. The activation functions of RELM, ELM and BP all choose the "Sigmoid" function: g(x)=1 / (1+exp(-x)), while the kernel function of SVM chooses the radial basis function. The input of the experimental data is always normalized to the range of [0, 1], while the output is normalized to the range of [-1, 1]...

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 belongs to the field of artificial intelligence, and relates to risk minimization and weighted least square theory. The invention discloses a regularized rapid neural network leaning method (Regularized Extreme Learning Machine, RELM). The method includes the following steps: (1) revising cost function, weighing empirical risks and structure risks, adjusting the ratio between the weighing empirical risks and the structure risks through parameters, and acquiring the best compromise of the two risks in the end; (2) revising model training, in order to reduce interference of outliers to models acquiring an anti-inference model, and adopting a method that error weighing different samples; (3) outputting weight calculating, first outputting initial value of the weight by using model training without the weight, acquiring each connecting weight and hidden layer threshold value of the neural network model by using the ELM technique, and finally acquiring the outputting value of the weight model.

Description

1. Technical field [0001] The invention belongs to the field of artificial intelligence, relates to the theory of risk minimization and weighted least squares, and discloses a regularized rapid learning method (Regularized Extreme Learning Machine, RELM) based on an extreme neural network model. 2. Background technology [0002] The reason why Single-hidden Layer Feedforward Neural Network (SLFN: Single-hidden Layer Feedforward Neural Network) can be widely used in many fields is because it has many advantages: (1) It has a strong learning ability and can approach complex nonlinear function; (2) It can solve problems that cannot be solved by traditional parameter methods. But on the other hand, the lack of fast learning methods also makes it unable to meet actual needs in many cases. [0003] For the learning ability of SLFN, many literatures have carried out in-depth research on the two input situations of compact input sets and infinite input sets. Hornik's research show...

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): G06N3/08
Inventor 邓万宇陈琳
Owner XIAN UNIV OF POSTS & TELECOMM
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