無(wú)限最大間隔線性判別投影模型

文偉; 曹雪菲; 陳渤; 韓勛; 張學(xué)峰; 王鵬輝; 劉宏偉

doi:10.11999/JEIT170256

無(wú)限最大間隔線性判別投影模型

doi: 10.11999/JEIT170256

基金項(xiàng)目:

國(guó)家杰出青年科學(xué)基金(61525105)，國(guó)家自然科學(xué)基金(61201292, 61322103, 61372132)，全國(guó)優(yōu)秀博士學(xué)位論文作者專項(xiàng)資金(FANEDD-201156)，陜西省自然科學(xué)基礎(chǔ)研究計(jì)劃(2016JQ6048)，航空科學(xué)基金(20142081009)，上海航天科技創(chuàng)新基金(SAST2015009)，航空電子系統(tǒng)射頻綜合仿真航空科技重點(diǎn)實(shí)驗(yàn)室基金

計(jì)量
- 文章訪問數(shù): 1457
- HTML全文瀏覽量: 194
- PDF下載量: 310
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2017-03-29
- 修回日期: 2017-09-21
- 刊出日期: 2017-12-19

Infinite Max-margin Linear Discriminant Projection Model

Funds:

The National Science Fund for Distinguished Young Scholars (61525105), The National Natural Science Foundation of China (61201292, 61322103, 61372132), The Program for New Century Excellent Talents in University (FANEDD-201156), The Natural Science Basic Research Plan in Shaanxi Province (2016JQ6048), The Avaation Science Fund (20142081009), Shanghai Aerospca Science, Technology Innovation Fund (SAST2015009), The Key Laboratory Fund of RF Integrated Laboratory in Avionics System

摘要

摘要: 針對(duì)具有多模分布結(jié)構(gòu)的高維數(shù)據(jù)的分類問題，該文提出一種無(wú)限最大間隔線性判別投影(iMMLDP)模型。與現(xiàn)有全局投影方法不同，模型通過聯(lián)合Dirichlet過程及最大間隔線性判別投影(MMLDP)模型將數(shù)據(jù)劃分為若干個(gè)局部區(qū)域，并在每一個(gè)局部學(xué)習(xí)一個(gè)最大邊界線性判別投影分類器。組合各局部分類器，實(shí)現(xiàn)全局非線性的投影與分類。iMMLDP模型利用貝葉斯框架聯(lián)合建模，將聚類、投影及分類器進(jìn)行聯(lián)合學(xué)習(xí)，可以有效發(fā)掘數(shù)據(jù)的隱含結(jié)構(gòu)信息，因而，可以較好地對(duì)非線性可分?jǐn)?shù)據(jù)，尤其是具有多模分布特性數(shù)據(jù)進(jìn)行分類。得益于非參數(shù)貝葉斯先驗(yàn)技術(shù)，可以有效避免模型選擇問題，即局部區(qū)域劃分?jǐn)?shù)量?；诜抡鏀?shù)據(jù)集、公共數(shù)據(jù)集及雷達(dá)實(shí)測(cè)數(shù)據(jù)集驗(yàn)證了所提方法的有效性。
- 最大間隔線性判別投影 /
- 非參數(shù)貝葉斯 /
- Dirichlet過程混合模型
Abstract: An infinite Max-Margin Linear Discriminant Projection (iMMLDP) model is developed to deal with the classification problem on multimodal distributed high-dimensional data. Different from global projection, iMMLDP divides the data into a set of local regions via Dirichlet Process (DP) mixture model and meanwhile learns a linear Max-Margin Linear Discriminant Projection (MMLDP) classifier in each local region. By assembling these local classifiers, a flexible nonlinear classifier is constructed. Under this framework, iMMLDP combines dimensionality reduction, clustering and supervised classification in a principled way, therefore, an underlying structure of the data could be uncovered. As a result, the model can handle the classification of data with global nonlinear structure, especially the data with multi-modally distributed structure. With the help of Bayesian nonparametric prior, the model selection problem (e.g. the number of local regions) can be avoided. The proposed model is implemented on synthesized and real-world data, including multi-modally distributed datasets and measured radar high range resolution profile (HRRP) data, to validate its efficiency and effectiveness.
- Max-Margin Linear Discriminant Projection (MMLDP) /
- Bayesian nonparametric /
- Dirichlet Process Mixture (DPM) model

HTML全文

參考文獻(xiàn)(26)

DUDA R, HART P, and STORK D. Pattern Classification (2nd ed.)[M]. New York, Wiley Interscience, 2000: 106-113.

郁道銀, 王悅行, 陳曉冬, 等. 基于隨機(jī)投影和稀疏表示的追蹤算法[J]. 電子與信息學(xué)報(bào), 2016, 38(7): 1602-1608. doi: 10.11999/JEIT151064.

YU Daoyin, WANG Yuexing, CHEN Xiaodong, et al. Visual tracking based on random projection and sparse representation[J]. Journal of Electronics Information Technology, 2016, 38(7): 1602-1608. doi: 10.11999/JEIT 151064.

SWETS D and WENG J. Using discriminant eigenfeatures for image retrieval[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(8): 831-836. doi: 10.1109/34.531802.

ETEMAD K and CHELLAPA R. Discriminant analysis for recognition of human face images[J]. Journal of the Optical Society of America A, 1997, 14(8): 1724-1733. doi: 10.1364 /JOSAA.14.001724.

FISHER R. The use of multiple measurements in taxonomic problems [J]. Annals of Eugenics, 1936, 7(2): 179188 doi: 10.1111/j.1469-1809.1936.tb02137.

CHEN B, ZHANG H, ZHANG X, et al. Max-margin discriminant projection via data augmentation[J]. IEEE Transactions on Knowledge and Data Engineering, 2015, 27(7): 1964-1976. doi: 10.1109/TKDE.2015.2397444.

NIKOLAOS G, VASILEIOS M, and IOANNIS K. Mixture Subclass discriminant analysis[J]. IEEE Signal Processing Letter, 2011, 18(5): 319-322. doi: 10.1109/LSP.2011. 2127474.

NIKOLAOS G, VASILEIOS M, and IOANNIS K. Mixture subclass discriminant analysis link to restricted Gaussian model and other generalizations[J]. IEEE Transactions on Neural Networks and Learning Systems, 2013, 24(1): 8-21. doi: 10.1109/TNNLS.2012.2216545.

郭繼昌, 張帆, 王楠. 基于Fisher約束和字典對(duì)的圖像分類 [J] 電子與信息學(xué)報(bào), 2017, 39(2): 270-277. doi: 10.11999/ JEIT160329.

GUO Jichang, ZHANG Fan, and WANG Nan. Image classification based on Fisher constraint and dictionary pair [J]. Journal of Electronics Information Technology, 2017, 39(2): 270-277. doi: 10.11999/JEIT160329.

GONEN M. Bayesian supervised dimensionality reduction [J]. IEEE Transactions on Cybernetics, 2013, 4(6): 2179-2189. doi: 10.1109/TCYB.2013.2245321.

SHAHBABA B and NEAL R. Nonlinear models using Dirichlet process mixtures[J]. The Journal of Machine Learning Research, 2009, 10(4): 1829-1850.

文偉, 曹雪菲, 張學(xué)峰, 等. 一種基于多極化散射機(jī)理的極化SAR圖像艦船目標(biāo)檢測(cè)方法[J]. 電子與信息學(xué)報(bào), 2017, 39(1): 103-109. doi: 10.11999/JEIT160204.

WEN Wei, CAO Xuefei, ZHANG Xuefeng, et al. PolSAR ship detection method based on multiple polarimetric scattering mechanisms[J]. Journal of Electronics Information Technology, 2017, 39(1): 103-109. doi: 10.11999 /JEIT160204.

POLSON N G and SCOTT S L. Data augmentation for support vector machines[J]. Bayesian Analysis, 2011, 6(1): 1-24. doi: 10.1214/11-BA601.

SETHURAMAN J. A constructive definition of Dirichlet priors[J]. Statistica Sinica, 1994, 4(2): 639-650.

HANNAH L A, BLEI D M, and POWELL W B. Dirichlet process mixtures of generalized linear models[J]. Journal of Machine Learning Research, 2011, 12: 1923-1953.

RIFKIN R and KLAUTAU A. In defense of one-vs-all classification[J]. Journal of Machine Learning Research, 2004, 5(1): 101-141.

DU L, LIU H W, BO Z, et al. Radar HRRP statistical recognition: Parametric model and model selection[J]. IEEE Transactions on Signal Processing, 2008, 56 (5): 1931-1943. doi: 10.1109/TSP.2007.912283.

張學(xué)峰, 陳渤, 王鵬輝, 等. 無(wú)限最大間隔Beta過程因子分析模型[J]. 西安電子科技大學(xué)學(xué)報(bào)(自然科學(xué)版), 2016, 43(3): 13-18. doi: 10.3969/j.issn.1001-2400.2016.03.003.

ZHANG Xuefeng, CHEN BO, WANG Penghui, et al. Infinite max-margin Beta process factor analysis model[J]. Journal of Xidian University (Natural Science), 2016, 43(3): 13-18. doi: 10.3969/j.issn.1001-2400.2016.03.003.

DU L, LIU H W, WANG P H, et al. Noise robust radar HRRP target recognition based on multitask factor analysis with small training data size[J]. IEEE Transactions on Signal Processing, 2012, 60(7): 3546-3559. doi: 10.1109/TSP. 2012.2191965.

張學(xué)峰, 陳渤, 王鵬輝, 等. 一種基于Dirichlet過程隱變量支撐向量機(jī)模型的目標(biāo)識(shí)別方法[J]. 電子與信息學(xué)報(bào), 2015, 37(1): 29-36. doi: 10.11999/JEIT140129.

ZHANG Xuefeng, CHEN Bo, WANG Penghui, et al. A

target recognition method based on Dirichlet process latent variable support vector machine model[J]. Journal of Electronics Information Technology, 2015, 37(1): 29-36. doi: 10.11999/JEIT140129.

相關(guān)文章

施引文獻(xiàn)

資源附件(0)

訪問統(tǒng)計(jì)