Adversarial Sampling for Active Learning

TL;DR

This paper introduces ASAL, a GAN-based active learning method that generates high-entropy samples and efficiently selects similar real samples from the pool, outperforming random sampling in multi-class tasks.

Contribution

ASAL is the first GAN-based active learning approach applicable to multi-class problems, offering high-quality sample generation and reduced computational complexity.

Findings

ASAL outperforms random sampling on multiple datasets.

ASAL has sub-linear runtime complexity.

ASAL performs best in certain data scenarios.

Abstract

This paper proposes asal, a new GAN based active learning method that generates high entropy samples. Instead of directly annotating the synthetic samples, ASAL searches similar samples from the pool and includes them for training. Hence, the quality of new samples is high and annotations are reliable. To the best of our knowledge, ASAL is the first GAN based AL method applicable to multi-class problems that outperforms random sample selection. Another benefit of ASAL is its small run-time complexity (sub-linear) compared to traditional uncertainty sampling (linear). We present a comprehensive set of experiments on multiple traditional data sets and show that ASAL outperforms similar methods and clearly exceeds the established baseline (random sampling). In the discussion section we analyze in which situations ASAL performs best and why it is sometimes hard to outperform random sample selection.