Dynamic depth confidence network analysis method

A technology of deep belief network and analysis method, applied in the direction of neural learning method, biological neural network model, instrument, etc., can solve the problems of lack of pertinence, do not consider some special classes, etc., and achieve the effect of improving accuracy

Active Publication Date: 2018-08-24
TONGJI UNIV
View PDF6 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The deep belief network has several obvious shortcomings: 1. When training the model, it strives to find the comprehensive optimal fit of all the data in the entire training set, without conside

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
  • Dynamic depth confidence network analysis method
  • Dynamic depth confidence network analysis method
  • Dynamic depth confidence network analysis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] In order to realize the object of the present invention, the training analysis strategy that the present invention provides is:

[0031] 1. Training stage:

[0032] In the training phase, a global network and a network pool containing a series of special networks need to be obtained.

[0033] First, a complete deep belief network is supervisedly trained with the entire training set, which is referred to as a global network in this invention. According to the principle of deep belief network, the purpose of the network is to maximize the difference between all categories, but not optimized for a specific category. Therefore, when the network analyzes a specific instance of a certain class, the output of some nodes in the network will instead become interference and affect the final classification result. For this reason, after training the global network, for each class, a special targeted network is trained based on the global network. The classification performance ...

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 the field of deep learning and relates to a dynamic network structure training analysis method based on a depth confidence network. The invention aims to overcome the severaldefects of the depth confidence network, and an overall-special dynamic network structure training analysis strategy from and based on the depth confidence network is provided. According to the strategy, a fine analysis process is introduced according to an analysis strategy of the brain when the human identifies an object. According to the strategy, two networks are generated in the training stage for the classification problem and comprise a global network and a plurality of special networks for specific categories. In the prediction stage, the output results of the two networks are comprehensively considered, so that the prediction of the whole model is more targeted, and the accuracy of the classification model is improved.

Description

technical field [0001] The invention relates to a dynamic network structure training analysis method based on a deep belief network in the field of deep learning. technical background [0002] ●Deep Belief Network [0003] Deep belief network (Deep belief network, DBN) is a probabilistic generation model in the field of deep learning. This model was proposed by Geoffrey Hinton in his paper "A Fast Learning Algorithm For Deep Belief Nets" published in 2006. It is usually used for classification . As opposed to traditional neural networks for discriminative models, deep belief networks are used to build a joint distribution between observations and labels. [0004] Theoretically, for the neural network structure, the more layers of hidden layers, the stronger the expressive power of the model. However, when the number of network layers gradually deepens, using the traditional gradient descent algorithm to optimize parameters will cause some problems. For example, when calc...

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): G06K9/62G06N3/08
CPCG06N3/082G06F18/21322G06F18/21324G06F18/24G06F18/214
Inventor 何良华胡仁杰罗裕隽莫文闻侍海峰刘洪宇王予沁任强刘晓洁蔡冠羽
Owner TONGJI UNIV
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