Name
Abstract | ||
|---|---|---|
We investigate a committee-based approach for active learning of real-valued functions. This is a variance-only strategy for selection of informative training data. As such it is shown to suffer when the model class is misspecified since the learner's bias is high. Conversely, the strategy outperforms passive selection when the model class is very expressive since active minimization of the variance avoids overfitting. |
Year | Venue | Keywords |
|---|---|---|
2007 | IDEAL | active minimization,variance avoids,active learning,informative training data,variance-only strategy,model class,real-valued function,passive selection,committee-based approach,value function |
Field | DocType | Volume |
Training set,Active learning,Active learning (machine learning),Pattern recognition,Regression,Computer science,Minification,Artificial intelligence,Overfitting,Machine learning | Conference | 4881 |
ISSN | ISBN | Citations |
0302-9743 | 3-540-77225-1 | 29 |
PageRank | References | Authors |
1.15 | 11 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Robert Burbidge | 1 | 35 | 2.26 |
| Jem J. Rowland | 2 | 61 | 8.22 |
| Ross D. King | 3 | 1774 | 194.85 |