摘要: The classical theory of pattern recognition assumes labeled examples appear according to unknown underlying class conditional probability distributions where the classes are picked randomly in a passive manner their priori probabilities. This paper presents experimental results for an incremental nearest-neighbor learning algorithm which actively selects samples from different querying rule as opposed amount improvement this query-based approach over batch depends on complexity Bayes rule.
core.ac.uk
computer.org
sci-hub.se 参考文章(14)
Vladimir Vapnik, Estimation of Dependences Based on Empirical Data: Springer Series in Statistics (Springer Series in Statistics) Springer-Verlag New York, Inc.. ,(1982)
Geoffrey W. Gates, The Reduced Nearest Neighbor Rule ,(1998)
Vladimir Naumovich Vapnik, Estimation of Dependences Based on Empirical Data ,(2010)
László Györfi, Luc Devroye, Gábor Lugosi, A Probabilistic Theory of Pattern Recognition ,(1996)
Ronald L. Rivest, Bonnie Eisenberg, On the sample complexity of pac-learning using random and chosen examples conference on learning theory. pp. 154- 162 ,(1990) , 10.5555/92571.92623
P. Hart, The condensed nearest neighbor rule (Corresp.) IEEE Transactions on Information Theory. ,vol. 14, pp. 515- 516 ,(1968) , 10.1109/TIT.1968.1054155
David A. Cohn, Neural network exploration using optimal experiment design Neural Networks. ,vol. 9, pp. 1071- 1083 ,(1996) , 10.1016/0893-6080(95)00137-9
Michael I. Shamos, Franco P. Preparata, Computational Geometry: An Introduction ,(1978)
G. Gates, The reduced nearest neighbor rule (Corresp.) IEEE Transactions on Information Theory. ,vol. 18, pp. 431- 433 ,(1972) , 10.1109/TIT.1972.1054809
Joseph O'Rourke, Computational geometry in C ,(1994)