Title :
On Kleinberg´s Stochastic Discrimination Procedure
Author :
Irle, Albrecht ; Kauschke, Jonas
Author_Institution :
Math. Inst., Univ. of Kiel, Kiel, Germany
fDate :
7/1/2011 12:00:00 AM
Abstract :
A new condition for high accuracy on test sets is given for the method of stochastic discrimination (SD), a pattern recognition method introduced by Kleinberg. This condition provides a simple explanation for the observed good generalization properties and overtraining-resistance. We also show that the method of SD remains valid if the original assumption of uniform distribution on a finite space for resampling is relaxed.
Keywords :
pattern classification; sampling methods; statistical distributions; stochastic processes; overtraining-resistance; pattern recognition method; resampling; stochastic discrimination procedure; uniform distribution; Accuracy; Algebra; Bagging; Computational modeling; Convergence; Pattern recognition; Training; Pattern recognition; stochastic discrimination.;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2010.225
