scholarly journals Transductive transfer learning via maximum margin criterion

2016 ◽  
Vol 23 (3) ◽  
pp. 1239-1250
Author(s):  
J. Tahmoresnezhad ◽  
S. Hashemi
Keyword(s):  
Transfer Learning ◽  
Maximum Margin ◽  
Maximum Margin Criterion

scholarly journals Rolling Bearing Fault Diagnosis Based on Sensitive Feature Transfer Learning and Local Maximum Margin Criterion under Variable Working Condition

2020 ◽  
Vol 2020 ◽  
pp. 1-34
Author(s):  
Shiyuan Liu ◽  
Xiao Yu ◽  
Xu Qian ◽  
Fei Dong
Keyword(s):  
Fault Diagnosis ◽  
Transfer Learning ◽  
Working Conditions ◽  
Local Maximum ◽  
Industrial Applications ◽  
Feature Reduction ◽  
Machine Learning Techniques ◽  
Mean Deviation ◽  
Maximum Margin ◽  
Maximum Margin Criterion

In real industrial scenarios, the working conditions of bearings are variable, and it is therefore difficult for data-driven diagnosis methods based on conventional machine-learning techniques to guarantee the desirable performance of diagnosis models, as the models assume that the distributions of both the training and testing data are the same. To enhance the performance of the fault diagnosis of bearings under different working conditions, a novel diagnosis framework inspired by feature extraction, transfer learning (TL), and feature dimensionality reduction is proposed in this work, and dual-tree complex wavelet packet transform (DTCWPT) is used for signal processing. Additionally, transferable sensitive feature selection by ReliefF and the sum of mean deviation (TSFSR) is proposed to reduce the redundant information of the original feature set, to select sensitive features for fault diagnosis, and to reduce the difference between the marginal distributions of the training and testing feature sets. Furthermore, a modified feature reduction method, the local maximum margin criterion (LMMC), is proposed to acquire low-dimensional mapping for high-dimensional feature spaces. Finally, bearing vibration signals collected from two test rigs are analyzed to demonstrate the adaptability, effectiveness, and practicability of the proposed diagnosis framework. The experimental results show that the proposed method can achieve high diagnosis accuracy and has significant potential benefits in industrial applications.

Contextual-distance Metric Based Laplacian Maximum Margin Criterion

2011 ◽  
Vol 36 (12) ◽  
pp. 1661-1673
Author(s):  
Jun GAO ◽  
Shi-Tong WANG ◽  
Xiao-Ming WANG
Keyword(s):  
Distance Metric ◽  
Maximum Margin ◽  
Maximum Margin Criterion

Feature extraction by structured stepwise nonparametric maximum margin criterion

2006 ◽  
Author(s):  
Yujie Zheng ◽  
Xiaojun Wu ◽  
Dongjun Yu ◽  
Jingyu Yang ◽  
Weidong Wang ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Maximum Margin ◽  
Maximum Margin Criterion

Early Image Fire-Detection Based on Maximum Margin Criterion

2014 ◽  
Vol 526 ◽  
pp. 324-329
Author(s):  
Jie Yuan ◽  
Hai Bing Hu ◽  
Wei Yuan ◽  
Yang Jia ◽  
Yong Ming Zhang
Keyword(s):  
Feature Extraction ◽  
Mathematical Description ◽  
Main Idea ◽  
Fire Detection ◽  
Color Model ◽  
Scatter Matrix ◽  
Maximum Margin ◽  
Maximum Margin Criterion ◽  
Early Image ◽  
The Difference

Nowadays as camera is applied widely, image fire detection becomes much popular. Many researchers are committed to analyze the RGB color model or even gray images. Actually they have some disadvantages. So this paper will present a new model based on Maximum Margin Criterion, a feature extraction criterion. As it is maximizing the difference of between-class scatter matrices and within-class scatter matrices, it does not depend on the nonsingularity of the within-class scatter matrix. First we will introduce the main idea and then give a mathematical description to apply the model to fire detection, with the algorithm we can calculate the result we need. At last we will put them into practice, use a database to do some experiments to present the performance of this method.

Sign in / Sign up

Export Citation Format

Share Document