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

Artificial neural network adjustment method and device

A technology of artificial neural network and adjustment method, applied in character and pattern recognition, instrument, calculation, etc., can solve the problems of adverse effects of classification results, insufficient discrimination between classes, and achieves the solution of adverse effects and uniform feature center angle distribution. Effect

Pending Publication Date: 2019-12-20
XILINX TECH BEIJING LTD
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention constrains the angle of the feature center by using a loss function, which can make the included angle between various feature centers more uniform, thereby well solving the problem of classification results due to insufficient discrimination between intra-class deviation and inter-class deviation. problems caused by adverse effects

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
  • Artificial neural network adjustment method and device
  • Artificial neural network adjustment method and device
  • Artificial neural network adjustment method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] Preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0028] The solution of the present application is applicable to various artificial neural networks (ANN), including deep neural network (DNN), recurrent neural network (RNN) and convolutional neural network (CNN). The following takes CNN as an example for some background explanation.

[0029] Basic concepts of CNN

[0030] CNNs achieve state-of-the-art performance on a wide range of vision-related tasks. To help understand the CNN-based cl...

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 an ANN (artificial neural network) adjustment method and device; the last full connection layer of the ANN is a classifier for classification, and the normalized weight of the layer represents various feature centers, and the method comprises the following steps: adjusting the ANN by using a first loss function for constraining an included angle of the feature centers; and completing the training of the ANN under the condition that the included angle distribution of each feature center tends to be uniform. According to the method, the angles of the feature centers are constrained by using the loss function, so that the included angles between the feature centers can be more uniform, and the problem of adverse effects on a classification result due to insufficient distinction degree of intra-class deviation and inter-class deviation can be well solved.

Description

technical field [0001] The present invention relates to deep learning, and in particular, to a method and device for adjusting artificial neural networks. Background technique [0002] In recent years, artificial neural network (ANN) has made significant progress in object detection, image classification and other fields. However, in engineering, unbalanced labeling data categories often occur. If there are many categories to be classified, such as face recognition tasks, the category imbalance of the above-mentioned labeled data will have a greater impact on the classification results due to insufficient discrimination between intra- and inter-class deviations. [0003] In order to solve the above problems, various improvement schemes have been proposed, such as data augmentation and local feature labeling. However, none of the above solutions can well solve the problem of inaccurate classification results caused by unbalanced categories of labeled data. [0004] In view...

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): G06K9/62
CPCG06V40/172G06F18/24G06F18/214
Inventor 高梓桁
Owner XILINX TECH BEIJING LTD
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