An Adversarial Regularisation for Semi-Supervised Training of Structured Output Neural Networks
This addresses the problem of reducing annotation costs for structured-output tasks like image segmentation, though it appears incremental as it builds on existing GAN frameworks.
The paper tackles semi-supervised training for structured-output neural networks by using a GAN-inspired discriminator to leverage unlabeled data, resulting in achieving the same performance as fully supervised training with half the annotations in image segmentation.
We propose a method for semi-supervised training of structured-output neural networks. Inspired by the framework of Generative Adversarial Networks (GAN), we train a discriminator network to capture the notion of a quality of network output. To this end, we leverage the qualitative difference between outputs obtained on the labelled training data and unannotated data. We then use the discriminator as a source of error signal for unlabelled data. This effectively boosts the performance of a network on a held out test set. Initial experiments in image segmentation demonstrate that the proposed framework enables achieving the same network performance as in a fully supervised scenario, while using two times less annotations.