288 pp. per issue
6 x 9, illustrated
2011 Impact Factor: 1.884
January 2004, Vol. 16, No. 1, Pages 115-137
Posted Online March 13, 2006.
© 2003 Massachusetts Institute of Technology
Asymptotic Properties of the Fisher Kernel
Max Planck Institute for Biological Cybernetics, 72076 Tübingen, Germany, and AIST Computational Biology Research Center, Koto-ku, Tokyo, 135-0064, Japan, email@example.comShotaro Akaho
AIST Neuroscience Research Institute, Tsukuba, 305-8568, Japan, firstname.lastname@example.orgMotoaki Kawanabe
Fraunhofer FIRST, 12489 Berlin, Germany, email@example.comKlaus-Robert Müller
Fraunhofer FIRST, 12489 Berlin, Germany, and University of Potsdam, 14482 Potsdam, Germany, firstname.lastname@example.org
This letter analyzes the Fisher kernel from a statistical point of view. The Fisher kernel is a particularly interesting method for constructing a model of the posterior probability that makes intelligent use of unlabeled data (i.e., of the underlying data density). It is important to analyze and ultimately understand the statistical properties of the Fisher kernel. To this end, we first establish sufficient conditions that the constructed posterior model is realizable (i.e., it contains the true distribution). Realizability immediately leads to consistency results. Subsequently, we focus on an asymptotic analysis of the generalization error, which elucidates the learning curves of the Fisher kernel and how unlabeled data contribute to learning. We also point out that the squared or log loss is theoretically preferable-because both yield consistent estimators-to other losses such as the exponential loss, when a linear classifier is used together with the Fisher kernel. Therefore, this letter underlines that the Fisher kernel should be viewed not as a heuristics but as a powerful statistical tool with well-controlled statistical properties.
, Yan Wu
. (2013) VE dimension induced by Bayesian networks over the boolean domain. Pattern Analysis and Applications
Online publication date: 19-Feb-2013.
, Sridhar Krishnan
, Aziz Guergachi
. (2012) Pattern Classification of Signals Using Fisher Kernels. Mathematical Problems in Engineering 2012
Online publication date: 1-Jan-2012.
, Yan Wu
. (2009) VC dimension and inner product space induced by Bayesian networks
. International Journal of Approximate Reasoning 50
Online publication date: 1-Jul-2009.
, Yan Wu
. (2009) Inner Product Space and Concept Classes Induced by Bayesian Networks. Acta Applicandae Mathematicae 106
Online publication date: 1-Jun-2009.
, H MAN
, A NEFIAN
. (2005) Face recognition based on multi-class mapping of Fisher scores. Pattern Recognition 38
Online publication date: 1-Jun-2005.
, H. Man
. (2005) Hybrid IMM/SVM approach for wavelet-domain probabilistic model based texture classification. IEE Proceedings - Vision, Image, and Signal Processing 152
Online publication date: 1-Jan-2005.