Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

43

JMLR
2008

favoriteEmaildiscussreport

79views more JMLR 2008»

Manifold Learning: The Price of Normalization

13 years 10 months ago

Manifold Learning: The Price of Normalization

Download jmlr.csail.mit.edu

We analyze the performance of a class of manifold-learning algorithms that find their output by minimizing a quadratic form under some normalization constraints. This class consists of Locally Linear Embedding (LLE), Laplacian Eigenmap, Local Tangent Space Alignment (LTSA), Hessian Eigenmaps (HLLE), and Diffusion maps. We present and prove conditions on the manifold that are necessary for the success of the algorithms. Both the finite sample case and the limit case are analyzed. We show that there are simple manifolds in which the necessary conditions are violated, and hence the algorithms cannot recover the underlying manifolds. Finally, we present numerical results that demonstrate our claims.

Yair Goldberg, Alon Zakai, Dan Kushnir, Yaacov Rit

Real-time Traffic

Algorithms | JMLR 2008 | Local Tangent Space Alignment | Manifolds |

claim paper

Related Content

» Learning speaker normalization using semisupervised manifold alignment

» 3D face tracking and expression inference from a 2D sequence using manifold learning

» Manifold Integration with Markov Random Walks

» Riemannian Manifold Learning

» Riemannian Manifold Learning for Nonlinear Dimensionality Reduction

» Regression on manifolds using kernel dimension reduction

» Learning a geometry integrated image appearance manifold from a small training set

» Learning and approximation by Gaussians on Riemannian manifolds

» Automatic 3D Facial Expression Analysis in Videos

Post Info
More Details (n/a)

Added	13 Dec 2010
Updated	13 Dec 2010
Type	Journal
Year	2008
Where	JMLR
Authors	Yair Goldberg, Alon Zakai, Dan Kushnir, Yaacov Ritov

Comments (0)