Investigating the Potential of Auxiliary-Classifier GANs for Image Classification in Low Data Regimes
This work addresses the challenge of limited training data for image classification, offering a potential all-in-one solution, though it appears incremental in its modifications to existing AC-GAN methods.
The paper tackled the problem of image classification in low data regimes by investigating Auxiliary-Classifier GANs (AC-GANs) as an integrated framework, achieving competitive performance with standard CNNs.
Generative Adversarial Networks (GANs) have shown promise in augmenting datasets and boosting convolutional neural networks' (CNN) performance on image classification tasks. But they introduce more hyperparameters to tune as well as the need for additional time and computational power to train supplementary to the CNN. In this work, we examine the potential for Auxiliary-Classifier GANs (AC-GANs) as a 'one-stop-shop' architecture for image classification, particularly in low data regimes. Additionally, we explore modifications to the typical AC-GAN framework, changing the generator's latent space sampling scheme and employing a Wasserstein loss with gradient penalty to stabilize the simultaneous training of image synthesis and classification. Through experiments on images of varying resolutions and complexity, we demonstrate that AC-GANs show promise in image classification, achieving competitive performance with standard CNNs. These methods can be employed as an 'all-in-one' framework with particular utility in the absence of large amounts of training data.