Publications
Detailed Information
Neural networks Ensemble for multi-dimensional classification : 다중 클래스 분류 문제를 위한 인공신경망 앙상블
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 문병로 | - |
dc.contributor.author | 박상철 | - |
dc.date.accessioned | 2017-07-14T02:34:23Z | - |
dc.date.available | 2017-07-14T02:34:23Z | - |
dc.date.issued | 2016-02 | - |
dc.identifier.other | 000000132149 | - |
dc.identifier.uri | https://hdl.handle.net/10371/122647 | - |
dc.description | 학위논문 (석사)-- 서울대학교 대학원 : 컴퓨터공학부, 2016. 2. 문병로. | - |
dc.description.abstract | Artificial neural network (ANN) is devised from human neural system in the field of machine learning. Several techniques including ensemble have been proposed to improve ANNs performance. In this thesis, we show three types of ensembles for classification task which approaches in the style of divide and conquer: pairwise, hierarchy, helper. The pairwise ensemble is using binary classifiers to solve one classification problem. The hierarchical ensemble is composed of networks which are independently trained to solve each small part of the problem. And the other ensemble method uses a traditional network that is assisted by a helper, trained to solve easy classification that is modified from original classification. The experiments are conducted with the MNIST database and the results show the problem and possibility of the style of divide and conquer in neural network. | - |
dc.description.tableofcontents | Chapter 1. Introduction 7
Chapter 2. Preliminaries 10 2.1 Artificial neural networks 10 2.2 Ensemble 11 2.3 Rectified linear unit 12 2.4 Dropout 13 2.5 ADADELTA 14 Chapter 3. Ensemble 15 3.1 Pairwise 15 3.2 Hierarchical ensemble 17 3.3 Helper 22 Chapter 4. Experiments 24 4.1 MNIST database 24 4.2 Experimental Setup 24 4.3 The result of pairwise ensemble 25 4.4 The result of hierarchical ensemble 27 4.5 The result of helper 28 Chapter 5. Conclusion 32 Bibliography 33 요약 36 | - |
dc.format | application/pdf | - |
dc.format.extent | 1179076 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject | Neural networks | - |
dc.subject | Ensemble | - |
dc.subject | Divide and conquer | - |
dc.subject.ddc | 621 | - |
dc.title | Neural networks Ensemble for multi-dimensional classification | - |
dc.title.alternative | 다중 클래스 분류 문제를 위한 인공신경망 앙상블 | - |
dc.type | Thesis | - |
dc.contributor.AlternativeAuthor | Sangchul Park | - |
dc.description.degree | Master | - |
dc.citation.pages | 36 | - |
dc.contributor.affiliation | 공과대학 컴퓨터공학부 | - |
dc.date.awarded | 2016-02 | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.