On Active Learning in Hierarchical Classification
Most of the existing active learning algorithms assume all the category labels as independent or consider them in a "Flat" structure. However, in reality, there are many applications in which the set of possible labels are often organized in a hierarchical structure. In this paper, the authors consider the problem of active learning when the categories are represented as a tree. Their goal is to exploit the structure information of the label tree in active learning to select the most informative samples to be labeled. They propose an algorithm that estimates the semantic space, embedding the category hierarchy. In this space, each category label is represented as a prototype and the uncertainty is measured using a variance-based fashion.