Neural network training method, device and electronic equipment

A neural network and training method technology, applied in the field of deep learning, to achieve the effect of sparse and reduced number

Active Publication Date: 2020-07-03
HORIZON ROBOTICS SHANGHAI ARTIFICIAL INTELLIGENCE TECH CO LTD
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In practical applications, there are two key bottlenecks in the calculation speed of the chip, one is the calculation speed, and the other is the data transmission delay

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
  • Neural network training method, device and electronic equipment
  • Neural network training method, device and electronic equipment
  • Neural network training method, device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Apparently, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described here.

[0026] Application overview

[0027] As mentioned above, by reducing the amount of data transferred, the computing speed of the chip can be increased. For example, the amount of transmitted data can be reduced by increasing the on-chip storage, thereby reducing the number of times of data transmission. However, the cost of on-chip storage is high, and increasing the on-chip storage will also increase the complexity of hardware design.

[0028] In view of the above technical problems, the inventors of the present application found that in the convolutional...

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

Disclosed are a neural network training method, a neural network training device and electronic equipment. The training method of the neural network includes: using sample data to train the first neural network to be trained; determining the index parameters of the first neural network in the training process; if the index parameters meet the preset conditions, determine and An updating manner corresponding to the preset condition; and updating parameters of a batch normalization layer in the first neural network based on the updating manner. In this way, the sparseness of the feature map output by the neural network is realized, thereby reducing the amount of transmitted data and further improving the calculation speed of the chip.

Description

technical field [0001] The present application relates to the field of deep learning, and more specifically, relates to a neural network training method, a neural network training device and electronic equipment. Background technique [0002] At present, with the rapid development of artificial intelligence technology, it is expected to provide more and more artificial intelligence services through terminals, such as smart phones, such as digital assistants and real-time translation, so there is a need for high-performance chips that can be used for neural network computing. growing demand. [0003] In practical applications, there are two key bottlenecks in the calculation speed of the chip, one is the calculation speed, and the other is the data transmission delay. In order to increase the calculation speed, methods such as increasing the number of calculation units and increasing the main frequency can be used. In order to reduce the delay of data transmission, you can ...

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 Patents(China)
IPC IPC(8): G06N3/08
CPCG06N3/084G06N3/045G06N20/20
Inventor 李智超高钰舒耿益锋罗恒
Owner HORIZON ROBOTICS SHANGHAI ARTIFICIAL INTELLIGENCE TECH CO LTD
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