The key idea behind active learning is that a machine learning algorithm can achieve
greater accuracy with fewer labeled training instances if it is allowed to choose the training …

Active learning aims to develop label-efficient algorithms by sampling the most
representative queries to be labeled by an oracle. We describe a pool-based semi …

In supervised learning, acquiring labeled training data for a predictive model can be very
costly, but acquiring a large amount of unlabeled data is often quite easy. Active learning is …

In this paper, we suggest a novel data-driven approach to active learning (AL). The key idea
is to train a regressor that predicts the expected error reduction for a candidate sample in a …

In all these cases, labels can be obtained, but only at a significant cost to the end user. An
important observation is that all records are not equally important from the perspective of …

Epistemic Uncertainty is a measure of the lack of knowledge of a learner which diminishes
with more evidence. While existing work focuses on using the variance of the Bayesian …

The recent breakthrough achieved by contrastive learning accelerates the pace for
deploying unsupervised training on real-world data applications. However, unlabeled data …

We introduce a model that learns active learning algorithms via metalearning. For a
distribution of related tasks, our model jointly learns: a data representation, an item selection …

Active learning can query valuable samples in an unlabeled sample pool for annotation,
thus building a more informative labeled dataset and reducing the annotation cost. However …

We present an agnostic active learning algorithm for any hypothesis class of bounded VC
dimension under arbitrary data distributions. Most previ-ous work on active learning either …