東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Competitive learning neural network ...

Ye, Qiang.

FindBook

Google Book

Amazon

博客來

Competitive learning neural network ensemble weighted by predicted performance.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Competitive learning neural network ensemble weighted by predicted performance./
作者:	Ye, Qiang.
面頁冊數:	88 p.
附註:	Source: Dissertation Abstracts International, Volume: 71-08, Section: A, page: 2690.
Contained By:	Dissertation Abstracts International71-08A.
標題:	Information Science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3417432
ISBN:	9781124147970

Competitive learning neural network ensemble weighted by predicted performance.
Ye, Qiang.

Competitive learning neural network ensemble weighted by predicted performance. - 88 p.

Source: Dissertation Abstracts International, Volume: 71-08, Section: A, page: 2690.

Thesis (Ph.D.)--University of Pittsburgh, 2010.

Ensemble approaches have been shown to enhance classification by combining the outputs from a set of voting classifiers. Diversity in error patterns among base classifiers promotes ensemble performance. Multi-task learning is an important characteristic for Neural Network classifiers. Introducing a secondary output unit that receives different training signals for base networks in an ensemble can effectively promote diversity and improve ensemble performance. Here a Competitive Learning Neural Network Ensemble is proposed where a secondary output unit predicts the classification performance of the primary output unit in each base network. The networks compete with each other on the basis of classification performance and partition the stimulus space. The secondary units adaptively receive different training signals depending on the competition. As the result, each base network develops "preference" over different regions of the stimulus space as indicated by their secondary unit outputs. To form an ensemble decision, all base networks' primary unit outputs are combined and weighted according to the secondary unit outputs. The effectiveness of the proposed approach is demonstrated with the experiments on one real-world and four artificial classification problems.

ISBN: 9781124147970Subjects--Topical Terms:

1017528
Information Science.

Competitive learning neural network ensemble weighted by predicted performance.
LDR:02326nam 2200301 4500 001 1401604
005 20111017084403.5
008 130515s2010 ||||||||||||||||| ||eng d
020 $a 9781124147970
035 $a (UMI)AAI3417432
035 $a AAI3417432
040 $a UMI $c UMI
100 1 $a Ye, Qiang. $3 718453
245 1 0 $a Competitive learning neural network ensemble weighted by predicted performance.
300 $a 88 p.
500 $a Source: Dissertation Abstracts International, Volume: 71-08, Section: A, page: 2690.
500 $a Adviser: Paul Munro.
502 $a Thesis (Ph.D.)--University of Pittsburgh, 2010.
520 $a Ensemble approaches have been shown to enhance classification by combining the outputs from a set of voting classifiers. Diversity in error patterns among base classifiers promotes ensemble performance. Multi-task learning is an important characteristic for Neural Network classifiers. Introducing a secondary output unit that receives different training signals for base networks in an ensemble can effectively promote diversity and improve ensemble performance. Here a Competitive Learning Neural Network Ensemble is proposed where a secondary output unit predicts the classification performance of the primary output unit in each base network. The networks compete with each other on the basis of classification performance and partition the stimulus space. The secondary units adaptively receive different training signals depending on the competition. As the result, each base network develops "preference" over different regions of the stimulus space as indicated by their secondary unit outputs. To form an ensemble decision, all base networks' primary unit outputs are combined and weighted according to the secondary unit outputs. The effectiveness of the proposed approach is demonstrated with the experiments on one real-world and four artificial classification problems.
520 $a Keywords: ensemble, diversity, neural networks, competitive learning, multi-task learning, bias and variance, classification
590 $a School code: 0178.
650 4 $a Information Science. $3 1017528
650 4 $a Artificial Intelligence. $3 769149
650 4 $a Computer Science. $3 626642
690 $a 0723
690 $a 0800
690 $a 0984
710 2 $a University of Pittsburgh. $3 958527
773 0 $t Dissertation Abstracts International $g 71-08A.
790 1 0 $a Munro, Paul, $e advisor
790 $a 0178
791 $a Ph.D.
792 $a 2010
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3417432