Generative Adversarial Active Learning

TL;DR

This paper introduces a novel active learning method that uses GANs to synthesize training instances, improving learning efficiency and outperforming traditional approaches in certain scenarios.

Contribution

It is the first to utilize GANs for active learning, adaptively generating queries to enhance learning speed and effectiveness.

Findings

Outperforms traditional pool-based active learning in some settings

Demonstrates effectiveness through numerical experiments

First to apply GANs in active learning context

Abstract

We propose a new active learning by query synthesis approach using Generative Adversarial Networks (GAN). Different from regular active learning, the resulting algorithm adaptively synthesizes training instances for querying to increase learning speed. We generate queries according to the uncertainty principle, but our idea can work with other active learning principles. We report results from various numerical experiments to demonstrate the effectiveness the proposed approach. In some settings, the proposed algorithm outperforms traditional pool-based approaches. To the best our knowledge, this is the first active learning work using GAN.