Discriminative Active Learning

TL;DR

Discriminative Active Learning (DAL) introduces a novel approach for batch active learning in neural networks by framing it as a binary classification problem, achieving competitive results with simplicity and broad applicability.

Contribution

DAL is a new batch active learning algorithm that treats example selection as a binary classification task, effective for large batch sizes and adaptable beyond classification.

Findings

DAL performs on par with state-of-the-art methods in medium and large batch sizes.

Existing methods are not clearly superior to uncertainty sampling at large batch sizes.

DAL is simple to implement and extend to other domains.

Abstract

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.