STAT SEMINAR :: Manifold Based Dimension Reduction and Statistical Properties
Manifold learning has attracted a lot of attention in both machine learning and statistics communities. I will give an overview and present many interesting applications. An appealing feature of many manifold learning algorithms is that their solutions correspond to linear invariant subspaces. Hence mathematical analysis can be implemented on the behavior of their solutions in a relatively common framework. I will describe some recent results regarding the worst case performance of manifold learning algorithms under additive noises.
- Workflow Status: Published
- Created By: Barbara Christopher
- Created: 10/08/2010
- Modified By: Fletcher Moore
- Modified: 10/07/2016