A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm
A Mahalanobis hyperellipsoidal learning machine class incremental learning algorithm is proposed. To each class sample, the hyperellipsoidal that encloses as many as possible and pushes the outlier samples away is trained in the feature space. In the process of incremental learning, only one subclas...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Hindawi Limited
2014-01-01
|
Series: | Abstract and Applied Analysis |
Online Access: | http://dx.doi.org/10.1155/2014/894246 |
id |
doaj-372f289e23404b03ab749ab824146fd5 |
---|---|
record_format |
Article |
spelling |
doaj-372f289e23404b03ab749ab824146fd52020-11-24T21:39:46ZengHindawi LimitedAbstract and Applied Analysis1085-33751687-04092014-01-01201410.1155/2014/894246894246A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning AlgorithmYuping Qin0Hamid Reza Karimi1Dan Li2Shuxian Lun3Aihua Zhang4College of Engineering, Bohai University, Jinzhou 121013, ChinaDepartment of Engineering, Faculty of Engineering and Science, The University of Agder, 4898 Grimstad, NorwayCollege of Mathematics and Physics, Bohai University, Jinzhou, ChinaNew Energy College, Bohai University, Jinzhou, ChinaCollege of Engineering, Bohai University, Jinzhou 121013, ChinaA Mahalanobis hyperellipsoidal learning machine class incremental learning algorithm is proposed. To each class sample, the hyperellipsoidal that encloses as many as possible and pushes the outlier samples away is trained in the feature space. In the process of incremental learning, only one subclassifier is trained with the new class samples. The old models of the classifier are not influenced and can be reused. In the process of classification, considering the information of sample’s distribution in the feature space, the Mahalanobis distances from the sample mapping to the center of each hyperellipsoidal are used to decide the classified sample class. The experimental results show that the proposed method has higher classification precision and classification speed.http://dx.doi.org/10.1155/2014/894246 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Yuping Qin Hamid Reza Karimi Dan Li Shuxian Lun Aihua Zhang |
spellingShingle |
Yuping Qin Hamid Reza Karimi Dan Li Shuxian Lun Aihua Zhang A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm Abstract and Applied Analysis |
author_facet |
Yuping Qin Hamid Reza Karimi Dan Li Shuxian Lun Aihua Zhang |
author_sort |
Yuping Qin |
title |
A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm |
title_short |
A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm |
title_full |
A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm |
title_fullStr |
A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm |
title_full_unstemmed |
A Mahalanobis Hyperellipsoidal Learning Machine Class Incremental Learning Algorithm |
title_sort |
mahalanobis hyperellipsoidal learning machine class incremental learning algorithm |
publisher |
Hindawi Limited |
series |
Abstract and Applied Analysis |
issn |
1085-3375 1687-0409 |
publishDate |
2014-01-01 |
description |
A Mahalanobis hyperellipsoidal learning machine class incremental learning algorithm is proposed. To each class sample, the hyperellipsoidal that encloses as many as possible and pushes the outlier samples away is trained in the feature space. In the process of incremental learning, only one subclassifier is trained with the new class samples. The old models of the classifier are not influenced and can be reused. In the process of classification, considering the information of sample’s distribution in the feature space, the Mahalanobis distances from the sample mapping to the center of each hyperellipsoidal are used to decide the classified sample class. The experimental results show that the proposed method has higher classification precision and classification speed. |
url |
http://dx.doi.org/10.1155/2014/894246 |
work_keys_str_mv |
AT yupingqin amahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT hamidrezakarimi amahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT danli amahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT shuxianlun amahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT aihuazhang amahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT yupingqin mahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT hamidrezakarimi mahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT danli mahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT shuxianlun mahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm AT aihuazhang mahalanobishyperellipsoidallearningmachineclassincrementallearningalgorithm |
_version_ |
1725929394627674112 |