[1]姜 峰,李 博,姚鸿勋,等.流形学习与基于线性耦合映射的流形对齐[J].智能系统学报,2010,5(6):476-481.
JIANG Feng,LI Bo,YAO Hong-xun,et al.Manifold learning and manifold alignmentbased on coupled linear projections[J].CAAI Transactions on Intelligent Systems,2010,5(6):476-481.
点击复制
JIANG Feng,LI Bo,YAO Hong-xun,et al.Manifold learning and manifold alignmentbased on coupled linear projections[J].CAAI Transactions on Intelligent Systems,2010,5(6):476-481.
流形学习与基于线性耦合映射的流形对齐
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
5
期数:
2010年第6期
页码:
476-481
栏目:
学术论文—人工智能基础
出版日期:
2010-12-25
- Title:
- Manifold learning and manifold alignmentbased on coupled linear projections
- 文章编号:
- 1673-4785(2010)06-0476-06
- Keywords:
- manifold learning; manifold alignment; coupled projection
- 分类号:
- TP18
- 文献标志码:
- A
- 摘要:
- 从一些具有代表性的经典流形学习方法的回顾来看，传统的流形学习主要处理来自单一流形的数据的降维问题.随着流形学习研究的不断深入，以多流形作为研究对象的流形学习问题逐步引起了研究者的注意.提出了一种基于线性耦合映射的流形对齐算法.算法克服非线性流形对齐算法不能够直接处理Outofsample数据的问题.同时，与已有的线性流形对齐算法相比，该算法不需要假设流形间满足仿射变换关系，因而能够更加灵活地处理一些比较实际的流形对齐问题.
- Abstract:
- Traditional manifold learning mainly deals with data dimensionality reduction problems from a single manifold. Through further development, manifold learning focusing on the multimanifold has drawn the attention of many scholars. An algorithm of manifold alignment based on coupled projections was proposed. The proposed method using the coupled linear projections dealt with an outofsample data set without retraining. Compared to other linear methods, it handled some real problems of manifold alignment with flexibility since the proposed method did not need any assumptions about the relationship of affine transformation between the manifolds.
- 参考文献/References:
-
［1］TENENBAUM J, SILVA V, LANGFORD J. A global geometric framework for nonlinear dimensionality reduction［J］. Science, 2000, 5500(290): 23192323.
［2］ROWEIS S, SAUL L. Nonlinear dimensionality reduction by locally linear embedding［J］. Science, 2000, 5500(290): 23232326.
［3］SAUL L, ROWEIS S. Think globally, fit locally: unsupervised learning of low dimensional manifolds［J］. The Journal of Machine Learning Research, 2003, 4: 119155.
［4］BELKIN M, NIYOGI P. Laplacian eigenmaps for dimensionality reduction and data representation［J］. Neural Computation, 2003, 15(6): 13731396.
［5］YAN S, XU D, ZHANG B, ZHANG H, YANG Q. Graph embedding and extensions: a general framework for dimensionality reduction［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(1): 4051.
［6］BENGIO Y, PAIEMENT J, VINCENT P, et al. Outofsample extensions for LIE, Isomap, MDS, eigenmaps, and spectral clustering［C］//Advances in Neural Information Processing Systems.Cambrige: MIT Press, 2004: 177184.
［7］HAM J, LEE D, SAUL L. Semisupervised alignment of manifolds［C］//Proceedings of the Annual Conference on Uncertainty in Artificial Intelligence. Edinburgh, UK， 2005: 120127.
［8］CONG Haifeng, PAN Chunhong, YANG Qing. A semisupervised framework for mapping data to the intrinsic manifold［C］//Tenth IEEE International Conference on Computer Vision 2005. ［S.l.］, 2005: 98105.
［9］XIONG Liang, WANG Fei, ZHANG Changshui. Semidefinite manifold alignment［J］. Lecture Notes in Computer Science, 2007, 4701: 773.
［10］RASMUSSEN C, WILLIAMS C. Gaussian processes for machine learning［M］. Cambrige: The MIT Press, 2006： 101107.
［11］WANG C, MAHADEVAN S. Manifold alignment using procrustes analysis［C］//Proceedings of the 25th International Conference on Machine Learning. Helsinki, Finland, 2008: 11201127.
［12］NENE S, NAYAR S, MURASE H. Columbia object image library (coil20)［R］. CUCS00696， Department of Computer Science, Columbia University.
［13］GEORGHIADES A, BELHUMEUR P, KRIEGMAN D. From few to many: illumination cone models for face recognition under variable lighting and pose［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(6): 643660.
备注/Memo
收稿日期：2009-12-05.
基金项目：黑龙江省自然科学基金资助项目(F200812).
通信作者：姜 峰.E-mail:fjiang@hit.edu.cn.
作者简介：
姜 峰，男，1978年生，硕士生导师.主要研究方向为模式识别、机器学习、图像处理等.
李 博，男，1980年生，博士研究生，主要研究方向为模式识别、机器学习、图像处理等.
姚鸿勋，女，1965年生，教授，博士生导师，主要研究方向为多媒体数据分析与理解、信息检索、视频监控、模式识别. 完成国家自然科学基金重点项目1项(评优)，国家自然科学基金1项(评优)，国家“863”计划重点项目子项目1项，“863”计划项目3项，“863”青年基金1项，信息产业部2项，国际合作4项.获省部级科技成果奖4项，其中一等奖2项，二等奖1项，三等奖1项，另获省级教学成果奖2项.已获国家发明专利4项，审理中5项.出版教材5部.已发表国内外学术论文160余篇，被SCI 检索32篇，EI 检索91篇，ISTP 检录61篇.
更新日期/Last Update:
2011-03-03