en
0.25
0.5
0.75
1.25
1.5
1.75
2
Aspects of Semi-Supervised and Active Learning in Conditional Random Fields
Published on Nov 30, 20113288 Views
Conditional random fields are among the state-of-the art approaches to structured output prediction, and the model has been adopted for various real-world problems. The supervised classification is ex
Related categories
Chapter list
Aspects of Semi-Supervised and Active Learning in Conditional Random Fields00:00
Outline - 100:17
Motivation01:38
Problem of Sequence Labeling: formalizations - 102:30
Problem of Sequence Labeling: formalizations - 202:50
Model of Conditional Random Fields03:17
Feature Functions04:04
Application: Phonetization task (NetTalk Corpus)05:14
Application: Named-Entity Recognition Task (CoNLL 2003)05:36
Outline - 206:00
Semi-Supervised Probabilistic Criterion06:08
Semi-Supervised Probabilistic Criterion Applied to CRF07:12
Semi-Supervised Criterion: Simulated Data07:47
Approximation of Marginal Probability of Observations08:53
Outline - 309:45
Motivation for Pool - Based Active Learning10:15
Active Learning - 111:17
Active Learning - 211:59
Conclusions and Perspectives13:48