Online choice of active learning algorithms
作者： Yoram Baram , Ran El-Yaniv , Kobi Luz
DOI:
关键词:
摘要: This paper is concerned with the question of how to online combine an ensemble active learners so as expedite learning progress during a pool-based session. We develop powerful master algorithm, based known competitive algorithm for multi-armed bandit problem and novel semi-supervised performance evaluation statistic. Taking containing two best algorithms new resulting empirically shown consistently perform almost well sometimes outperform in on range classification problems.
参考文章(28)
Nello Cristianini, Alex J. Smola, Colin Campbell, Query Learning with Large Margin Classifiers international conference on machine learning. pp. 111- 118 ,(2000)
Nicholas Roy, Andrew McCallum, Toward Optimal Active Learning through Sampling Estimation of Error Reduction international conference on machine learning. pp. 441- 448 ,(2001)
David Cohn, Greg Schohn, Less is More: Active Learning with Support Vector Machines international conference on machine learning. pp. 839- 846 ,(2000)
Yoav Freund, H. Sebastian Seung, Eli Shamir, Naftali Tishby, Selective Sampling Using the Query by Committee Algorithm Machine Learning. ,vol. 28, pp. 133- 168 ,(1997) , 10.1023/A:1007330508534
G. Rätsch, T. Onoda, K.-R. Müller, Soft Margins for AdaBoost Machine Learning. ,vol. 42, pp. 287- 320 ,(2001) , 10.1023/A:1007618119488
Bayes point machines Journal of Machine Learning Research. ,vol. 1, pp. 245- 279 ,(2001) , 10.1162/153244301753683717
Vladimir Vapnik, Isabelle Guyon, Nada Matic, Discovering informative patterns and data cleaning knowledge discovery and data mining. pp. 181- 203 ,(1996)
Nello Cristianini, J Shawe-Taylor, An introduction to Support Vector Machines Cambridge University Press (2000). ,(2000)
Maytal Saar-Tsechansky, Foster Provost, Active Sampling for Class Probability Estimation and Ranking Social Science Research Network. ,(2001)
John Shawe-Taylor, Nello Cristianini, Further results on the margin distribution Proceedings of the twelfth annual conference on Computational learning theory - COLT '99. pp. 278- 285 ,(1999) , 10.1145/307400.307470