Deep Bayesian Active Learning with Image Data

TL;DR

This paper integrates Bayesian deep learning into active learning to effectively select informative image data, overcoming challenges of small data updates and uncertainty estimation, leading to improved performance on MNIST and skin cancer datasets.

Contribution

It introduces a practical active learning framework using Bayesian convolutional neural networks for high-dimensional image data, a novel approach in this context.

Findings

Significant improvement over existing active learning methods.

Effective application to both MNIST and skin cancer diagnosis datasets.

Demonstrates feasibility of Bayesian deep learning in high-dimensional active learning.

Abstract

Even though active learning forms an important pillar of machine learning, deep learning tools are not prevalent within it. Deep learning poses several difficulties when used in an active learning setting. First, active learning (AL) methods generally rely on being able to learn and update models from small amounts of data. Recent advances in deep learning, on the other hand, are notorious for their dependence on large amounts of data. Second, many AL acquisition functions rely on model uncertainty, yet deep learning methods rarely represent such model uncertainty. In this paper we combine recent advances in Bayesian deep learning into the active learning framework in a practical way. We develop an active learning framework for high dimensional data, a task which has been extremely challenging so far, with very sparse existing literature. Taking advantage of specialised models such as Bayesian convolutional neural networks, we demonstrate our active learning techniques with image data, obtaining a significant improvement on existing active learning approaches. We demonstrate this on both the MNIST dataset, as well as for skin cancer diagnosis from lesion images (ISIC2016 task).