Paper Info
Title
Rademacher Complexities and Bounding the Excess Risk in Active Learning
Abstract
Sequential algorithms of active learning based on the estimation of the level sets of the empirical risk are discussed in the paper. Localized Rademacher complexities are used in the algorithms to estimate the sample sizes needed to achieve the required accuracy of learning in an adaptive way. Probabilistic bounds on the number of active examples have been proved and several applications to binary classification problems are considered.
Year
DOI
Venue
2010
10.5555/1756006.1953014
Journal of Machine Learning Research
Keywords
Field
DocType
sample size,probabilistic bound,required accuracy,active example,localized rademacher complexity,rademacher complexities,empirical risk,level set,classification problem,sequential algorithm,excess risk,active learning,binary classification
Active learning,Binary classification,Pattern recognition,Active learning (machine learning),Rademacher complexity,Level set,Artificial intelligence,Probabilistic logic,Sample size determination,Mathematics,Machine learning,Bounding overwatch
Journal
Volume
ISSN
Citations 
11,
1532-4435
33
PageRank 
References 
Authors
1.45
11
1
Name
Order
Citations
PageRank
Vladimir Koltchinskii1899.61