Margin Based Active Learning

We present a framework for margin based active learning of linear separators. We instantiate it for a few important cases, some of which have been previously considered in the literature. We analyze the effectiveness of our framework both in the realizable case and in a specific noisy setting relate...

Full description

Saved in:
Bibliographic Details
Published inLearning Theory Vol. 4539; pp. 35 - 50
Main Authors Balcan, Maria-Florina, Broder, Andrei, Zhang, Tong
Format Book Chapter
LanguageEnglish
Published Germany Springer Berlin / Heidelberg 2007
Springer Berlin Heidelberg
SeriesLecture Notes in Computer Science
Subjects
Active Learning
Computational Learn Theory
Sample Complexity
Unit Ball
Unlabeled Data
Online AccessGet full text
ISBN9783540729259
3540729259
ISSN0302-9743
1611-3349
DOI10.1007/978-3-540-72927-3_5

Cover

More Information
Summary:We present a framework for margin based active learning of linear separators. We instantiate it for a few important cases, some of which have been previously considered in the literature. We analyze the effectiveness of our framework both in the realizable case and in a specific noisy setting related to the Tsybakov small noise condition.
ISBN:9783540729259
3540729259
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-540-72927-3_5