An Optimized Sparse Auto-encoder Network Based on Feature Clustering

doi:10.3969/j.issn.0372-2112.2018.05.003

您当前的位置：

首页 >

文章列表页 >

An Optimized Sparse Auto-encoder Network Based on Feature Clustering

更新时间：2025-07-08

- An Optimized Sparse Auto-encoder Network Based on Feature Clustering
- Acta Electronica Sinica Vol. 46, Issue 5, Pages: 1041-1046(2018)
- 作者机构：
  
  中国地质大学数学与物理学院,湖北,武汉,430074
- 作者简介：
- 基金信息：
  
  Youth Fund of National Natural Science Foundation of China (No.61601417);Program for New Century Excellent Talents in University of Ministry of Education of China (No.NCET-13-1011)
- DOI：10.3969/j.issn.0372-2112.2018.05.003
  CLC： TP391
- Published Online：25 May 2018，
  
  Published：2018
- 稿件说明：
移动端阅览
An Optimized Sparse Auto-encoder Network Based on Feature Clustering[J]. Acta Electronica Sinica, 2018, 46(5): 1041-1046.
DOI：

An Optimized Sparse Auto-encoder Network Based on Feature Clustering[J]. Acta Electronica Sinica, 2018, 46(5): 1041-1046. DOI： 10.3969/j.issn.0372-2112.2018.05.003.

摘要

稀疏自编码网络在自然语言、图像处理等领域都取得了显著效果.已有的研究表明增加网络提取的特征个数可以优化稀疏自编码网络的处理效果，同时该操作将导致网络训练耗时过长.为尽可能减少网络的训练时间，本文提出了一种基于特征聚类的稀疏自编码快速算法.本算法首先根据K均值聚类最优数确定本质特征的个数，再由网络训练得到本质特征，并通过旋转扭曲增加特征的多样性，使网络处理效果得到提升的同时，减少网络训练耗间.实验在标准的手写体识别数据库MNIST和人脸数据库CMU-PIE上进行，结果表明本文所提算法能在保证网络正确率有所提升的同时，大幅度缩短网络训练耗时.

Abstract

The method of deep sparse auto-encoder networks has achieved state-of-art performance in the fields of image processing and natural language processing.It's been proved that higher accuracy of deep sparse auto-encoder networks is obtained by the increase of features' number

however

it also leads to a longer training time.In this paper

an optimized sparse auto-encoder networks which based on feature clustering is been presented for both classification accuracy enhancement and training time decreasing.The proposed method first get the number of substantive features by optimizing k-means clustering.Then initialize the network with that number and obtain the substantive features by training again the network.Finally the improvement of feature varieties is achieved by rotation and distortion of the substantive features.In the experiments

the improvement of classification accuracy and reduction of training time is verified by comparing the performance of optimized sparse auto-encoder with normal sparse auto-encoder in the basic dataset MNIST and CMU-PIE.

关键词

Keywords

references

Views

266

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on the Generation of Adversarial Samples Based on the Attribution of Principal Features

Review on Image Scene Classification Technology

Abnormal Event Detection Based on Deep Learning

Survey of Object Detection Based on Deep Learning

Related Author

WANG Shuo

XU Ru-zhi

GUAN Zhi-tao

TIAN Yan-ling

ZHANG Wei-tong

ZHANG Qie-shi

LU Gang

WU Xiao-jun

Related Institution

School of Control and Computer Engineering， North China Electric Power University

Key Laboratory of Modern Teaching Technology, Ministry of Education, Shaanxi Normal University

School of Computer Science, Shaanxi Normal University

Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences

Key Laboratory of Modern Teaching Technology Ministry of Education Shaanxi Normal University Xi'an Shaanxi China

⁰