Title :
Robustness in statistical pattern recognition under “contaminations” of training samples
Author :
Kharin, Yu. ; Zhuk, E.
Author_Institution :
Dept. of Math. Modeling & Data Anal., Belarusian State Univ., Minsk, Russia
Abstract :
The problems of synthesis and analysis of statistical pattern recognition algorithms are considered for the situations, when traditional model assumptions about training samples are disturbed. The estimates of robustness measures for classical decision rules and new robust decision rules are constructed for the case of “contaminated” training samples
Keywords :
statistical analysis; algorithm analysis; algorithm synthesis; contaminated training samples; robustness measures; statistical pattern recognition robustness; Algorithm design and analysis; Bayesian methods; Contamination; Loss measurement; Pattern analysis; Pattern recognition; Pollution measurement; Robust stability; Robustness; Statistics;
Conference_Titel :
Pattern Recognition, 1994. Vol. 2 - Conference B: Computer Vision & Image Processing., Proceedings of the 12th IAPR International. Conference on
Conference_Location :
Jerusalem
Print_ISBN :
0-8186-6270-0
DOI :
10.1109/ICPR.1994.576996
