Discriminative Active Learning
This work addresses active learning for neural networks in batch settings, but it is incremental as it builds on existing paradigms and shows competitive rather than groundbreaking results.
The paper tackles the problem of batch mode active learning for neural networks by proposing Discriminative Active Learning (DAL), which frames active learning as a binary classification task to select examples that make labeled and unlabeled sets indistinguishable; experiments on image classification show DAL performs on par with state-of-the-art methods for medium and large batch sizes, while also revealing that no current methods clearly outperform uncertainty sampling for large batches.
We propose a new batch mode active learning algorithm designed for neural networks and large query batch sizes. The method, Discriminative Active Learning (DAL), poses active learning as a binary classification task, attempting to choose examples to label in such a way as to make the labeled set and the unlabeled pool indistinguishable. Experimenting on image classification tasks, we empirically show our method to be on par with state of the art methods in medium and large query batch sizes, while being simple to implement and also extend to other domains besides classification tasks. Our experiments also show that none of the state of the art methods of today are clearly better than uncertainty sampling when the batch size is relatively large, negating some of the reported results in the recent literature.