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

Constructive neural network method based on knowledge cascading correlation

A neural network and cascade technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as low flexibility and scalability, waste of manpower, etc., to achieve good adaptability and flexibility, fast The effect of learning speed

Inactive Publication Date: 2017-06-30
CHINA UNIV OF MINING & TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method brings a great waste of manpower, and the flexibility and scalability are very low

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
  • Constructive neural network method based on knowledge cascading correlation
  • Constructive neural network method based on knowledge cascading correlation
  • Constructive neural network method based on knowledge cascading correlation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0012] The realization of a knowledge-based cascade correlation constructive neural network method of the present invention consists of an initialization stage, an output stage and an input stage. Among them, the output stage and the input stage will decide whether to jump to another stage for learning according to their respective judgment standards.

[0013] Combine below figure 1 The three implementation stages of a knowledge-based cascaded correlation constructive neural network method of the present invention are introduced in detail.

[0014] Initialization phase:

[0015] Step 1, before starting training, initialize the network connection weights.

[0016] Output stage:

[0017] Step 1, use optimization algorithms such as backpropagation algorithm or fast propagation algorithm to train output weights. In this step, the function F to be optimized is the error sum of squares function on all training samples p and output nodes o:

[0018] F=∑ o ∑ p (V o,p -T o,p )...

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 provides a constructive neural network method based on knowledge cascading correlation, wherein the method relates to the field of artificial neural network, machine learning and artificial intelligence. Particularly the invention relates to cascading of learned knowledge with the neural network through relativity, and furthermore the neural network is constructively reconstructed based on cascading of the knowledge, thereby realizing a purpose of quickly and flexibly constructing the neural network. According to the method, a network structure and weight are adjusted according to training data in a frontward propagating neural network. The method mainly comprises two periods, namely an input period and an output period. In the output period, all weights which enter an output mode is optimized through a training algorithm so that the output error of a target network is reduced continuously. In the input period, a new hidden layer node or knowledge network is added into the target network. A candidate network which is finally selected for being added into the candidate network can best correlate an output error of the target network.

Description

technical field [0001] The invention relates to a knowledge-based cascade correlation constructive neural network method, which relates to the fields of artificial neural network, machine learning and artificial intelligence. In particular, it involves cascading the learned knowledge into the neural network through correlation, and constructively transforming the neural network on this basis, so as to achieve the purpose of quickly and flexibly constructing the neural network. Background technique [0002] At present, the neural networks that are widely used at home and abroad must determine the network structure before training and learning. This preset network structure may not be suitable for a given learning task in practical applications, for example: the network structure is too simple and the neural network does not have enough learning ability; or the network structure is too complex, although it can meet the needs of the learning task , but the complex structure br...

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): G06N3/04G06N3/08
CPCG06N3/08G06N3/045
Inventor 丁世飞孟令恒徐晓赵星宇张健张楠
Owner CHINA UNIV OF MINING & TECH
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