Improving Generalization with Active Learning
Machine Learning - Special issue on structured connectionist systems
Selective Sampling Using the Query by Committee Algorithm
Machine Learning
ALT '01 Proceedings of the 12th International Conference on Algorithmic Learning Theory
ICML '06 Proceedings of the 23rd international conference on Machine learning
Performance thresholding in practical text classification
CIKM '06 Proceedings of the 15th ACM international conference on Information and knowledge management
A bound on the label complexity of agnostic active learning
Proceedings of the 24th international conference on Machine learning
Hierarchical sampling for active learning
Proceedings of the 25th international conference on Machine learning
Importance weighted active learning
ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning
Theoretical foundations of active learning
Theoretical foundations of active learning
COLT'07 Proceedings of the 20th annual conference on Learning theory
Minimax Bounds for Active Learning
IEEE Transactions on Information Theory
On multiple-instance learning of halfspaces
Information Processing Letters
Querying discriminative and representative samples for batch mode active learning
Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining
Active labeling application applied to food-related object recognition
Proceedings of the 5th international workshop on Multimedia for cooking & eating activities
| Hi-index | 5.23 |
An active learner has a collection of data points, each with a label that is initially hidden but can be obtained at some cost. Without spending too much, it wishes to find a classifier that will accurately map points to labels. There are two common intuitions about how this learning process should be organized: (i) by choosing query points that shrink the space of candidate classifiers as rapidly as possible; and (ii) by exploiting natural clusters in the (unlabeled) data set. Recent research has yielded learning algorithms for both paradigms that are efficient, work with generic hypothesis classes, and have rigorously characterized labeling requirements. Here we survey these advances by focusing on two representative algorithms and discussing their mathematical properties and empirical performance.