Publications
Detailed Information
Multiple Kernel Learning by Gap Maximization between Classes in RKHS : 커널 공간에서 클래스 샘플들간의 상대적 거리 최대화로 구현한 커널 학습 방법론
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 곽노준 | - |
dc.contributor.author | John Yang | - |
dc.date.accessioned | 2017-07-19T10:57:06Z | - |
dc.date.available | 2017-07-19T10:57:06Z | - |
dc.date.issued | 2016-02 | - |
dc.identifier.other | 000000133632 | - |
dc.identifier.uri | https://hdl.handle.net/10371/133212 | - |
dc.description | 학위논문 (석사)-- 서울대학교 융합과학기술대학원 : 융합과학기술대학원 융합과학부 지능형융합시스템학 전공, 2016. 2. 곽노준. | - |
dc.description.abstract | By implicitly maximizing the gap between classes in the reproducing kernel Hilbert
space (RKHS), a multiple kernel learning (MKL) is formulated as a linear programming in this paper. For each sample, my method tries to enforce the distance between intra-class and inter-class samples in RKHS to be as distant as possible. n my method, each training sample imposes at most r constraints for the linear programming where r is the number of different kernel types. Unlike previous methods of multiple kernel learning, the proposed method does not need a large amount of computations. My method is compared with various methods of MKL to prove its efficiency of finding a good kernel mixture parameter. | - |
dc.description.tableofcontents | Chapter 1 Introduction 1
Chapter 2 RelatedWorks 5 2.1 Alignment-based MKL (ABMKL) 6 2.2 Centered-alignment-based MKL (CABMKL) 8 2.3 MKL, Simple MKL (SimpleMKL), Generalized MKL (GMKL) 9 2.4 The Group Lasso-based MKL (GLMKL) 12 2.5 Non-linear MKL (NLMKL) 12 2.6 Localized MKL (LMKL) 13 Chapter 3 MKL by Gap-Maximization (MKL-GM) 16 3.1 Motivation: Kernel Target Alignment 16 3.2 MKL-GM for the same type of kernels 18 3.3 MKL-GM for different types of kernels 23 Chapter 4 Experiments 27 4.1 Toy example: Two-spiral data 27 4.2 FERET face database 31 4.3 Protein Fold Prediction 33 4.4 Pendigits Digit Recognition 36 4.5 Clatech-101 dataset 38 Chapter 5 Conclusion and Future Work 40 Bibliography 42 | - |
dc.format | application/pdf | - |
dc.format.extent | 479814 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 융합과학기술대학원 | - |
dc.subject | Multiple Kernel Learning | - |
dc.subject | Reproducing Kernel Hilbert Space | - |
dc.subject | Kernel Method | - |
dc.subject | Kernel Alignment | - |
dc.subject | Kernel Trick | - |
dc.subject.ddc | 620 | - |
dc.title | Multiple Kernel Learning by Gap Maximization between Classes in RKHS | - |
dc.title.alternative | 커널 공간에서 클래스 샘플들간의 상대적 거리 최대화로 구현한 커널 학습 방법론 | - |
dc.type | Thesis | - |
dc.contributor.AlternativeAuthor | 양한열 | - |
dc.description.degree | Master | - |
dc.citation.pages | 47 | - |
dc.contributor.affiliation | 융합과학기술대학원 융합과학부 | - |
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.