Active learning

Active learning is a form of supervised machine learning in which a learning algorithm is able to interactively query the user (or some other information source) to obtain the desired outputs at new data points. In statistics literature it is sometimes also called optimal experimental design.[1][2]

There are situations in which unlabeled data is abundant but manually labeling it is expensive. In such a scenario, learning algorithms can actively query the user/teacher for labels. This type of iterative supervised learning is called active learning. Since the learner chooses the examples, the number of examples to learn a concept can often be much lower than the number required in normal supervised learning. With this approach, there is a risk that the algorithm be overwhelmed by uninformative examples.


Voici la référence de l'article de POPL:

Synthesis of biological models from mutation experiments

http://dl.acm.org/citation.cfm?id=2429125


Erick