Semi-Supervised Learning with GANs: Revisiting Manifold Regularization
This work improves semi-supervised learning for image classification by making it easier to implement, though it is incremental.
The paper tackled the problem of semi-supervised learning by using GANs to approximate manifold regularization, achieving state-of-the-art results on CIFAR-10 with a simpler implementation.
GANS are powerful generative models that are able to model the manifold of natural images. We leverage this property to perform manifold regularization by approximating the Laplacian norm using a Monte Carlo approximation that is easily computed with the GAN. When incorporated into the feature-matching GAN of Improved GAN, we achieve state-of-the-art results for GAN-based semi-supervised learning on the CIFAR-10 dataset, with a method that is significantly easier to implement than competing methods.