A Dimensionality Reduction Matlab Toolbox: Too much of a good thing ?

[If the links to the toolbox does not work, you may want to look for another location in the comments section of this entry ]


Laurens van der Maaten just released a large matlab toolbox featuring 25 dimensionality reduction techniques [Updated: Dec 1, 2008]. And then you can use PRTools to do your own classification. Be careful what you wish for :-)

The list of techniques readily implemented:

Principal Component Analysis (PCA)
Simple PCA
Probabilistic PCA
Linear Discriminant Analysis (LDA)
Multidimensional scaling (MDS)
Isomap
Landmark Isomap
Local Linear Embedding (LLE)
Laplacian Eigenmaps
Hessian LLE
Local Tangent Space Alignment (LTSA)
Conformal Eigenmaps (extension of LLE)
Maximum Variance Unfolding (extension of LLE)
Fast Maximum Variance Unfolding (FastMVU)
Kernel PCA
Generalized Discriminant Analysis (GDA)
Diffusion maps
Stochastic Neighbor Embedding (SNE)
Neighborhood Preserving Embedding (NPE)
Locality Preserving Projection (LPP)
Linear Local Tangent Space Alignment (LLTSA)
Stochastic Proximity Embedding (SPE)
Multilayer autoencoders (training by RBM + backpropagation or by an evolutionary algorithm)
Local Linear Coordination (LLC)
Manifold charting
Coordinated Factor Analysis (CFA)