SIMILAR: Submodular Information Measures Based Active Learning In Realistic Scenarios

TL;DR

SIMILAR introduces a unified active learning framework using submodular information measures, effectively handling realistic challenges like class imbalance, out-of-distribution data, and redundancy, and significantly improves performance on image classification tasks.

Contribution

The paper presents SIMILAR, a novel active learning framework based on submodular information measures, extending applicability to realistic scenarios and demonstrating superior performance.

Findings

Outperforms existing methods by 5-18% on rare classes.

Achieves 5-10% improvement on out-of-distribution data.

Scalable to large real-world datasets.

Abstract

Active learning has proven to be useful for minimizing labeling costs by selecting the most informative samples. However, existing active learning methods do not work well in realistic scenarios such as imbalance or rare classes, out-of-distribution data in the unlabeled set, and redundancy. In this work, we propose SIMILAR (Submodular Information Measures based actIve LeARning), a unified active learning framework using recently proposed submodular information measures (SIM) as acquisition functions. We argue that SIMILAR not only works in standard active learning, but also easily extends to the realistic settings considered above and acts as a one-stop solution for active learning that is scalable to large real-world datasets. Empirically, we show that SIMILAR significantly outperforms existing active learning algorithms by as much as ~5% - 18% in the case of rare classes and ~5% - 10% in the case of out-of-distribution data on several image classification tasks like CIFAR-10, MNIST, and ImageNet. SIMILAR is available as a part of the DISTIL toolkit: "https://github.com/decile-team/distil".