Name
Abstract | ||
|---|---|---|
We introduce confidence-weighted linear classifiers, which add parameter confidence information to linear classifiers. Online learners in this setting update both classifier parameters and the estimate of their confidence. The particular online algorithms we study here maintain a Gaussian distribution over parameter vectors and update the mean and covariance of the distribution with each instance. Empirical evaluation on a range of NLP tasks show that our algorithm improves over other state of the art online and batch methods, learns faster in the online setting, and lends itself to better classifier combination after parallel training. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1145/1390156.1390190 | ICML |
Keywords | Field | DocType |
classifier combination,online setting,parameter vector,parameter confidence information,art online,confidence-weighted linear classification,gaussian distribution,particular online,linear classifier,nlp task,classifier parameter,online algorithm | Online algorithm,Pattern recognition,Computer science,Gaussian,Artificial intelligence,Classifier (linguistics),Linear classifier,Machine learning,Covariance | Conference |
Citations | PageRank | References |
121 | 6.79 | 9 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Mark Dredze | 1 | 3092 | 176.22 |
| Koby Crammer | 2 | 5252 | 466.86 |
| Fernando Pereira | 3 | 17717 | 2124.79 |