SoQal: Selective Oracle Questioning in Active Learning
arXiv:2004.10468
Abstract
Large sets of unlabelled data within the healthcare domain remain underutilized. Active learning offers a way to exploit these datasets by iteratively requesting an oracle (e.g. medical professional) to label instances. This process, which can be costly and time-consuming is overly-dependent upon an oracle. To alleviate this burden, we propose SoQal, a questioning strategy that dynamically determines when a label should be requested from an oracle. We perform experiments on five publically-available datasets and illustrate SoQal's superiority relative to baseline approaches, including its ability to reduce oracle label requests by up to 35%. SoQal also performs competitively in the presence of label noise: a scenario that simulates clinicians' uncertain diagnoses when faced with difficult classification tasks.
References in corpus (8)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Predicting Cardiovascular Risk Factors from Retinal Fundus Photographs using Deep Learning
- Bayesian Active Learning for Classification and Preference Learning
- Learning Active Learning from Data
- Active Learning from Weak and Strong Labelers
- O-MedAL: Online Active Deep Learning for Medical Image Analysis
- Active Learning from Imperfect Labelers
- Icebreaker: Element-wise Active Information Acquisition with Bayesian Deep Latent Gaussian Model