Supervised Dimension Reduction of Intrinsically Low-Dimensional Data Export

@article{Vlassis_02, abstract = {doi: 10.1162/089976602753284491 High-dimensional data generated by a system with limited degrees of freedom are often constrained in low-dimensional manifolds in the original space. In this article, we investigate dimension-reduction methods for such intrinsically low-dimensional data through linear projections that preserve the manifold structure of the data. For intrinsically one-dimensional data, this implies projecting to a curve on the plane with as few intersections as possible. We are proposing a supervised projection pursuit method that can be regarded as an extension of the single-index model for nonparametric regression. We show results from a toy and two robotic applications.}, address = {Cambridge, MA, USA}, author = {Vlassis, N. and Motomura, Y. and Kr\"{o}se, B.}, citeulike-article-id = {893457}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=639715}, citeulike-linkout-1 = {http://dx.doi.org/10.1162/089976602753284491}, citeulike-linkout-2 = {http://www.mitpressjournals.org/doi/abs/10.1162/089976602753284491}, day = {1}, doi = {10.1162/089976602753284491}, issn = {0899-7667}, journal = {Neural Computation}, keywords = {dimension\_reduction, learning, lwpr}, month = {January}, number = {1}, pages = {191--215}, posted-at = {2009-11-04 11:19:17}, priority = {2}, publisher = {MIT Press}, title = {Supervised Dimension Reduction of Intrinsically Low-Dimensional Data}, url = {http://dx.doi.org/10.1162/089976602753284491}, volume = {14}, year = {2002} }

TY - JOUR ID - Vlassis_02 L3 - citeulike-article-id:893457 N2 - doi: 10.1162/089976602753284491 High-dimensional data generated by a system with limited degrees of freedom are often constrained in low-dimensional manifolds in the original space. In this article, we investigate dimension-reduction methods for such intrinsically low-dimensional data through linear projections that preserve the manifold structure of the data. For intrinsically one-dimensional data, this implies projecting to a curve on the plane with as few intersections as possible. We are proposing a supervised projection pursuit method that can be regarded as an extension of the single-index model for nonparametric regression. We show results from a toy and two robotic applications. IS - 1 JF - Neural Computation SN - 0899-7667 AD - Cambridge, MA, USA EP - 215 TI - Supervised Dimension Reduction of Intrinsically Low-Dimensional Data PB - MIT Press VL - 14 SP - 191 KW - dimension_reduction KW - learning KW - lwpr AU - Vlassis, N AU - Motomura, Y AU - Kröse, B PY - 2002/01/01/ UR - http://dx.doi.org/10.1162/089976602753284491 ER -