A Meta-Learning Approach to One-Step Active Learning

TL;DR

This paper introduces a meta-learning framework for active learning that learns to select data points to label efficiently in a single shot, improving label efficiency in costly labeling scenarios.

Contribution

It proposes a novel meta-learning approach to learn active-learning strategies directly, unlike traditional heuristic or theoretical methods.

Findings

Encouraging experimental results demonstrate the effectiveness of the learned strategies.

The approach outperforms some baseline methods in label efficiency.

The method is applicable in pool-based active learning settings.

Abstract

We consider the problem of learning when obtaining the training labels is costly, which is usually tackled in the literature using active-learning techniques. These approaches provide strategies to choose the examples to label before or during training. These strategies are usually based on heuristics or even theoretical measures, but are not learned as they are directly used during training. We design a model which aims at \textit{learning active-learning strategies} using a meta-learning setting. More specifically, we consider a pool-based setting, where the system observes all the examples of the dataset of a problem and has to choose the subset of examples to label in a single shot. Experiments show encouraging results.