Deep learning neural network evolution method and apparatus, medium and computer device

A neural network and deep learning technology, applied in neural learning methods, biological neural network models, physical realization, etc., can solve the problem of low accuracy of deep learning neural networks, achieve the effect of correcting deviations and improving accuracy

Active Publication Date: 2018-05-08
SOUTH CHINA NORMAL UNIVERSITY
View PDF15 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to provide a deep learning neural network evolution method, device, medium and computer equipment that can improve the accuracy of the deep learning neural network for the problem of low accuracy of the traditional deep learning neural network

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
  • Deep learning neural network evolution method and apparatus, medium and computer device
  • Deep learning neural network evolution method and apparatus, medium and computer device
  • Deep learning neural network evolution method and apparatus, medium and computer device

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0113] Specific embodiment one: the preset time period is from the earliest use time to the present; the preset threshold is 0; the group number threshold is 3.

[0114] Compare the difference between the output data and the corresponding real result data in each data group in the evolutionary knowledge base, count the number of groups whose difference value is greater than 0 from the earliest use time to now, if the number of groups is greater than 3, use the evolutionary knowledge base The input data and the corresponding real result data in the data group whose difference value is greater than 0 from the earliest use time to the present are used to train the currently trained deep learning neural network. The evolutionary knowledge base has the following datasets:

[0115] On January 1st, the input data is face image A, the output data is 1 (representing three faces), and the corresponding real result data is 1 (representing three faces). The difference between the output ...

specific Embodiment 2

[0137] Specific embodiment 2: the preset time period is the last 5 times; the preset threshold is 0; the group number threshold is 2.

[0138] Compare the difference between the output data and the corresponding real result data in each data group in the evolutionary knowledge base, count the number of groups whose difference value is greater than 0 in the last 5 times, if the number of groups is greater than 3, use the latest 5 in the evolutionary knowledge base The input data and the corresponding real result data in the data group whose intra-time difference value is greater than 0 are used to train the currently trained deep learning neural network. The evolutionary knowledge base has the following datasets:

[0139] On January 1st, the input data is face image A, the output data is 1 (representing three faces), and the corresponding real result data is 1 (representing three faces). The difference between the output data and the corresponding real result data is equal to ...

specific Embodiment 3

[0159] Specific embodiment three: the preset time period is from the earliest use time to the present; the preset threshold is 0; the group number threshold=(the total number of data groups in the evolutionary knowledge base in the preset time period) / 4.

[0160] Compare the difference between the output data and the corresponding real result data in each data group in the evolutionary knowledge base, count the number of groups whose difference value is greater than 0 from the earliest use time to now, if the number of groups is greater than the threshold of the number of groups, use The input data and the corresponding real result data in the data group whose difference value is greater than 0 from the earliest use time to the present in the evolutionary knowledge base are used to train the currently trained deep learning neural network. The evolutionary knowledge base has the following datasets:

[0161] On January 1st, the input data is face image A, the output data is 1 (r...

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 a deep learning neural network evolution method and apparatus, a medium and a computer device. The method includes the steps of inputting input data into a trained deep learning neural network that is called to obtain output data, recording usage time and obtaining corresponding real result data; selecting the input data, output data, usage time and real result data duringuse according to preset sampling rules; forming data groups with the selected input data, corresponding output data, real result data and usage time; calculating a difference value between the outputdata and the real result data in the data groups, and conducting statistics on the number of groups, in which the difference value is greater than a preset threshold, in all the data groups within apreset period of time; and if the number of groups is greater than or equal to a stored threshold of the number of groups, training the trained deep learning neural network based on the input data andreal result data of the difference groups to obtain an evolved deep learning neural network. In this way, the deviation of the deep learning neural network can be corrected and the accuracy is increased.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a deep learning neural network evolution method, device, medium and computer equipment. Background technique [0002] Neural network is an algorithmic mathematical model for distributed parallel information processing, which is often used in intelligent machine recognition. The deep learning neural network is a neural network obtained after deep learning through sample data training and passing the test. Among them, deep learning is divided into supervised learning and unsupervised learning. [0003] Usually the deep learning neural network does not change after the deep learning neural network is obtained, that is, it does not change during the use of the trained deep learning neural network. However, the training and testing data of the deep learning neural network is limited after all, and the accuracy rate cannot be guaranteed, and as the input data changes over ti...

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): G06N3/08G06N3/063
CPCG06N3/086G06N3/065
Inventor 朱定局
Owner SOUTH CHINA NORMAL UNIVERSITY
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