Sign inSign up
Optimally Combining Classifiers Using Unlabeled Data thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Optimally Combining Classifiers Using Unlabeled Data

Published on Aug 20, 20151691 Views

Akshay Balsubramani

We develop a worst-case analysis of aggregation of classifier ensembles for binary classification. The task of predicting to minimize error is formulated as a game played over a given set of unlabeled

COLT 2015 - Paris

Related categories

Active LearningComputational Learning TheoryOn-line LearningReinforcement LearningSemi-supervised Learning

Chapter list

Optimally combining classifiers using unlabeled data00:00
Setup00:05
An illustration - 100:43
An illustration - 201:00
An illustration - 301:17
An illustration - 401:58
Possible to beat ERM and majority vote?02:14
Better than ERM and majority vote - 102:46
Better than ERM and majority vote - 203:33
Formulation03:56
Optimal Strategies - 104:35
Optimal Strategies - 205:26