Vanishing Twin GAN: How training a weak Generative Adversarial Network can improve semi-supervised image classification
This addresses a specific issue in semi-supervised learning for image classification, offering an incremental improvement for scenarios where class similarity reduces accuracy.
The paper tackles the problem of semi-supervised image classification where unknown classes share characteristics with known classes, which can cause GANs to generalize and hurt performance; it proposes the Vanishing Twin GAN, which improves classification by training a weak GAN alongside a regular GAN.
Generative Adversarial Networks can learn the mapping of random noise to realistic images in a semi-supervised framework. This mapping ability can be used for semi-supervised image classification to detect images of an unknown class where there is no training data to be used for supervised classification. However, if the unknown class shares similar characteristics to the known class(es), GANs can learn to generalize and generate images that look like both classes. This generalization ability can hinder the classification performance. In this work, we propose the Vanishing Twin GAN. By training a weak GAN and using its generated output image parallel to the regular GAN, the Vanishing Twin training improves semi-supervised image classification where image similarity can hurt classification tasks.