Sciweavers

10 search results - page 1 / 2
» PAC Classification based on PAC Estimates of Label Class Dis...
Sort
View
CORR
2006
Springer
74views Education» more  CORR 2006»
13 years 4 months ago
PAC Classification based on PAC Estimates of Label Class Distributions
A standard approach in pattern classification is to estimate the distributions of the label classes, and then to apply the Bayes classifier to the estimates of the distributions in...
Nick Palmer, Paul W. Goldberg
JMLR
2006
109views more  JMLR 2006»
13 years 4 months ago
Some Discriminant-Based PAC Algorithms
A classical approach in multi-class pattern classification is the following. Estimate probability distributions that generated the observations for each label class, and then labe...
Paul W. Goldberg
COLT
2005
Springer
13 years 10 months ago
A PAC-Style Model for Learning from Labeled and Unlabeled Data
Abstract. There has been growing interest in practice in using unlabeled data together with labeled data in machine learning, and a number of different approaches have been develo...
Maria-Florina Balcan, Avrim Blum
COLT
2008
Springer
13 years 6 months ago
Does Unlabeled Data Provably Help? Worst-case Analysis of the Sample Complexity of Semi-Supervised Learning
We study the potential benefits to classification prediction that arise from having access to unlabeled samples. We compare learning in the semi-supervised model to the standard, ...
Shai Ben-David, Tyler Lu, Dávid Pál
AI
2002
Springer
13 years 4 months ago
Learning cost-sensitive active classifiers
Most classification algorithms are "passive", in that they assign a class label to each instance based only on the description given, even if that description is incompl...
Russell Greiner, Adam J. Grove, Dan Roth