Bayesian Conditional Generative Adverserial Networks
This work addresses generative modeling for unsupervised, supervised, and semi-supervised learning, but it appears incremental as it extends GANs to a Bayesian framework without specifying a major breakthrough.
The paper tackles the problem of traditional GANs using deterministic generators by proposing Bayesian Conditional GANs (BC-GANs) with a random generator function, and it reports that BC-GANs outperform state-of-the-art methods in experiments.
Traditional GANs use a deterministic generator function (typically a neural network) to transform a random noise input $z$ to a sample $\mathbf{x}$ that the discriminator seeks to distinguish. We propose a new GAN called Bayesian Conditional Generative Adversarial Networks (BC-GANs) that use a random generator function to transform a deterministic input $y'$ to a sample $\mathbf{x}$. Our BC-GANs extend traditional GANs to a Bayesian framework, and naturally handle unsupervised learning, supervised learning, and semi-supervised learning problems. Experiments show that the proposed BC-GANs outperforms the state-of-the-arts.