Task-Aware Variational Adversarial Active Learning

Often, labeling large amount of data is challenging due to high labeling cost limiting the application domain of deep learning techniques. Active learning (AL) tackles this by querying the most informative samples to be annotated among unlabeled pool. Two promising directions for AL that have been recently explored are task-agnostic approach to select data points that are far from the current labeled pool and task-aware approach that relies on the perspective of task model. Unfortunately, the former does not exploit structures from tasks and the latter does not seem to well-utilize overall data distribution. Here, we propose task-aware variational adversarial AL (TA-VAAL) that modifies task-agnostic VAAL, that considered data distribution of both label and unlabeled pools, by relaxing task learning loss prediction to ranking loss prediction and by using ranking conditional generative adversarial network to embed normalized ranking loss information on VAAL. Our proposed TA-VAAL outperforms state-of-the-arts on various benchmark datasets for classifications with balanced / imbalanced labels as well as semantic segmentation and its task-aware and task-agnostic AL properties were confirmed with our in-depth analyses.

1. Introduction

Deep learning has achieved remarkable performance in various computer vision tasks such as classification [18], [13], object detection [27], [26], and semantic segmentation [19], [4] due to massive datasets with annotations such as ImageNet for image classification [7] and PASCAL VOC for classification, detection, segmentation [9]. Obtaining good annotations is challenging and has often been a large-scale project. Moreover, there are often cases where labeling massive amount of data is even more challenging or infeasible due to high labeling cost such as labeling by experts [8] or long labeling time per large-scale sample such as videos [1] or pathology images [3]. Labeling cost seems to be a factor to limit the scope of applicability of deep learning to more research areas and more institutes with less labeling budget.