FINE: Fisher Information Nonparametric Embedding Export

@article{citeulike:5940281, abstract = {We consider the problems of clustering, classification, and visualization of high-dimensional data when no straightforward euclidean representation exists. In this paper, we propose using the properties of information geometry and statistical manifolds in order to define similarities between data sets using the Fisher information distance. We will show that this metric can be approximated using entirely nonparametric methods, as the parameterization and geometry of the manifold is generally unknown. Furthermore, by using multidimensional scaling methods, we are able to reconstruct the statistical manifold in a low-dimensional euclidean space; enabling effective learning on the data. As a whole, we refer to our framework as Fisher Information Nonparametric Embedding (FINE) and illustrate its uses on practical problems, including a biomedical application and document classification.}, address = {Los Alamitos, CA, USA}, author = {Carter, Kevin M. and Raich, Raviv and Finn, William G. and Hero, Alfred O.}, citeulike-article-id = {5940281}, citeulike-linkout-0 = {http://doi.ieeecomputersociety.org/10.1109/TPAMI.2009.67}, citeulike-linkout-1 = {http://dx.doi.org/10.1109/TPAMI.2009.67}, citeulike-linkout-2 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4815255}, day = {17}, doi = {10.1109/TPAMI.2009.67}, issn = {0162-8828}, journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, keywords = {embedding, fisher, information, manifolds}, month = {April}, number = {11}, pages = {2093--2098}, posted-at = {2009-11-10 05:09:23}, priority = {3}, publisher = {IEEE Computer Society}, title = {FINE: Fisher Information Nonparametric Embedding}, url = {http://dx.doi.org/10.1109/TPAMI.2009.67}, volume = {31}, year = {2009} }

TY - JOUR ID - citeulike:5940281 L3 - citeulike-article-id:5940281 N2 - We consider the problems of clustering, classification, and visualization of high-dimensional data when no straightforward euclidean representation exists. In this paper, we propose using the properties of information geometry and statistical manifolds in order to define similarities between data sets using the Fisher information distance. We will show that this metric can be approximated using entirely nonparametric methods, as the parameterization and geometry of the manifold is generally unknown. Furthermore, by using multidimensional scaling methods, we are able to reconstruct the statistical manifold in a low-dimensional euclidean space; enabling effective learning on the data. As a whole, we refer to our framework as Fisher Information Nonparametric Embedding (FINE) and illustrate its uses on practical problems, including a biomedical application and document classification. IS - 11 JF - IEEE Transactions on Pattern Analysis and Machine Intelligence SN - 0162-8828 AD - Los Alamitos, CA, USA EP - 2098 TI - FINE: Fisher Information Nonparametric Embedding PB - IEEE Computer Society VL - 31 SP - 2093 KW - embedding KW - fisher KW - information KW - manifolds AU - Carter, Kevin AU - Raich, Raviv AU - Finn, William AU - Hero, Alfred PY - 2009/04/17/ UR - http://dx.doi.org/10.1109/TPAMI.2009.67 ER -