Learning Bundle Manifold by Double Neighborhood Graphs

In this paper, instead of the ordinary manifold assumption, we introduced the bundle manifold assumption that imagines data points lie on a bundle manifold Under this assumption, we proposed an unsupervised algorithm, named as Bundle Manifold Embedding (BME), to embed the bundle manifold into low dimensional space In BME, we construct two neighborhood graphs that one is used to model the global metric structure in local neighborhood and the other is used to provide the information of subtle structure, and then apply the spectral graph method to obtain the low-dimensional embedding Incorporating some prior information, it is possible to find the subtle structures on bundle manifold in an unsupervised manner Experiments conducted on benchmark datasets demonstrated the feasibility of the proposed BME algorithm, and the difference compared with ISOMAP, LLE and Laplacian Eigenmaps.

[1]  Lawrence K. Saul,et al.  Think Globally, Fit Locally: Unsupervised Learning of Low Dimensional Manifold , 2003, J. Mach. Learn. Res..

[2]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[3]  Daniel D. Lee,et al.  Semisupervised alignment of manifolds , 2005, AISTATS.

[4]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.

[5]  Mikhail Belkin,et al.  Laplacian Eigenmaps for Dimensionality Reduction and Data Representation , 2003, Neural Computation.

[6]  D. Donoho,et al.  Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[7]  Ann B. Lee,et al.  Geometric diffusions as a tool for harmonic analysis and structure definition of data: diffusion maps. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[8]  Kilian Q. Weinberger,et al.  Unsupervised Learning of Image Manifolds by Semidefinite Programming , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[9]  Mikhail Belkin,et al.  Semi-Supervised Learning on Riemannian Manifolds , 2004, Machine Learning.

[10]  H. Zha,et al.  Principal manifolds and nonlinear dimensionality reduction via tangent space alignment , 2004, SIAM J. Sci. Comput..

[11]  Rajesh P. N. Rao,et al.  Learning Lie Groups for Invariant Visual Perception , 1998, NIPS.

[12]  Daniel D. Lee,et al.  Separating Pose and Expression in Face Images: A Manifold Learning Approach , 2007 .

[13]  Jiří Matas,et al.  Computer Vision - ECCV 2004 , 2004, Lecture Notes in Computer Science.

[14]  Fei Wang,et al.  Label Propagation through Linear Neighborhoods , 2008, IEEE Trans. Knowl. Data Eng..

[15]  Matthew Brand,et al.  Charting a Manifold , 2002, NIPS.