Active Learning techniques are used to tackle learning problems where obtaining training labels is costly. In this work we use Meta-Active Learning to learn to select a subset of samples from a pool of unsupervised input for further annotation. This scenario is called Static Pool-based Meta- Active Learning. We propose to extend existing approaches by performing the selection in a manner that, unlike previous works, can handle the selection of each sample based on the whole selected subset.
updated: Fri Nov 01 2019 12:08:47 GMT+0000 (UTC)
published: Fri Nov 01 2019 12:08:47 GMT+0000 (UTC)