[R] [R-pkgs] "tsne" package for t-SNE dimensionality reduction

[Justin Donaldson]

I'd like to announce the availability of the "tsne" package.

It provides the T distributed Stocastic Neighbor Embedding algorithm by van
der Maaten and Hinton:


   1. http://ict.ewi.tudelft.nl/~lvandermaaten/t-SNE.html

   2.

   L.J.P. van der Maaten and G.E. Hinton. Visualizing High-Dimensional Data
   Using t-SNE. Journal of Machine Learning Research 9(Nov):2579-2605, 2008.
   [ PDF<http://ict.ewi.tudelft.nl/%7Elvandermaaten/t-SNE_files/vandermaaten08a.pdf>]
[ Supplemental
   Material<http://ict.ewi.tudelft.nl/%7Elvandermaaten/t-SNE_files/SupplementalMaterial.pdf>(24MB)
]
   3.

   L.J.P. van der Maaten. Learning a Parametric Embedding by Preserving
   Local Structure. In Proceedings of the Twelfth International Conference
   on Artificial Intelligence and Statistics (AISTATS), JMLR W&CP 5:384-391,
   2009. [ PDF<http://ict.ewi.tudelft.nl/%7Elvandermaaten/t-SNE_files/Paper.pdf>]

It has a few utility/convenience parameters for tweaking the algorithm and
providing feedback.  Be sure to try the example to get an idea of how it
works:

require(tsne)
?tsne

It's currently much slower than the "official" implementations by van der
Maaten.  If you're interested in providing some tips on optimized C/Fortan
code, let me know.


Best,
-Justin

-- 
Justin Donaldson
PhD Candidate, Informatics
Indiana University
http://www.scwn.net
aim: iujjd
twitter: jjdonald

	[[alternative HTML version deleted]]

_______________________________________________
R-packages mailing list
R-packages at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-packages