Semi-supervised Conditional GANs
This work addresses the challenge of generating data conditioned on attributes with limited labeled data, which is an incremental improvement for machine learning applications in domains like image synthesis.
The paper tackles the problem of building conditional generative models in a semi-supervised setting by introducing a semi-supervised GAN (SS-GAN) that uses stacked discriminators to learn distributions from labeled and unlabeled data, resulting in significantly better performance compared to existing models.
We introduce a new model for building conditional generative models in a semi-supervised setting to conditionally generate data given attributes by adapting the GAN framework. The proposed semi-supervised GAN (SS-GAN) model uses a pair of stacked discriminators to learn the marginal distribution of the data, and the conditional distribution of the attributes given the data respectively. In the semi-supervised setting, the marginal distribution (which is often harder to learn) is learned from the labeled + unlabeled data, and the conditional distribution is learned purely from the labeled data. Our experimental results demonstrate that this model performs significantly better compared to existing semi-supervised conditional GAN models.