Non-Local Manifold Tangent Learning - Archive ouverte HAL Access content directly
Conference Papers Year : 2004

Non-Local Manifold Tangent Learning


We claim and present arguments to the effect that a large class of manifold learning algorithms that are essentially local and can be framed as kernel learning algorithms will suffer from the curse of dimensionality, at the dimension of the true underlying manifold. This observation suggests to explore non-local manifold learning algorithms which attempt to discover shared structure in the tangent planes at different positions. A criterion for such an algorithm is proposed and experiments estimating a tangent plane prediction function are presented, showing its advantages with respect to local manifold learning algorithms: it is able to generalize very far from training data (on learning handwritten character image rotations), where a local non-parametric method fails.
Fichier principal
Vignette du fichier
Non_Local_Manifold_Tangent_Learning.pdf (117.4 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01575349 , version 1 (18-08-2017)


  • HAL Id : hal-01575349 , version 1


Yoshua Bengio, Martin Monperrus. Non-Local Manifold Tangent Learning. Advances in Neural Information Processing Systems, 2004, Vancouver, Canada. pp.129-136. ⟨hal-01575349⟩
33 View
202 Download


Gmail Facebook X LinkedIn More