论文信息 - Incremental Hessian Locally Linear Embedding algorithm

Incremental Hessian Locally Linear Embedding algorithm

In this paper we propose an incremental version of Hessian locally linear embedding (HLLE) on the basis of incremental locally linear embedding (LLE) generalizations. The main idea behind our algorithm is to produce a lower dimension representation of a high-dimensional manifold such that the significant characteristics of the dataset are preserved while adapting to newly added points arriving to the dataset. Our experimental results verify how the new projection of points, along with the additional points, produces a good fit to the original manifold.

Amr M. Youssef | A. Ben Hamza | Osama Abdel-Mannan

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

[2] D. Donoho,et al. Hessian Eigenmaps : new locally linear embedding techniques for high-dimensional data , 2003 .

[3] Matti Pietikäinen,et al. Selection of the Optimal Parameter value for the Locally Linear Embedding Algorithm , 2002, FSKD.

[4] 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.

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

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

[7] Matti Pietikäinen,et al. Incremental Locally Linear Embedding Algorithm , 2005, SCIA.

[8] R. Bishop,et al. Geometry of Manifolds , 1964 .