pdf

Research on Neural Network Learning Methods

  • 2013-09-21
  • 1.65MB
  • Points it Requires : 2

  Single-hidden Layer Feedforward Neural Network (SLFN) has been widely used in pattern recognition, automatic control and data mining. However, the speed of traditional learning methods is far from meeting the actual needs, which has become the main bottleneck restricting its development. There are two main reasons for this situation: (1) the traditional error back propagation method (BP) is mainly based on the idea of ​​gradient descent and requires multiple iterations; (2) all parameters of the network need to be iteratively determined during the training process. Therefore, the algorithm has a large amount of calculation and search space. To solve the above problems, a fast learning method (RELM) is proposed based on the one-shot learning idea of ​​ELM and the structural risk minimization theory, which avoids multiple iterations and local minimums and has good generalization, robustness and controllability. Experiments show that the comprehensive performance of RELM is better than that of ELM, BP and SVM.  

unfold

You Might Like

Uploader
froglucky
 

Recommended ContentMore

Popular Components

Just Take a LookMore

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号
×