Graphical Generative Adversarial Networks
This work addresses the challenge of generating structured data for machine learning applications, representing an incremental advancement by integrating existing techniques.
The authors tackled the problem of modeling structured data by proposing Graphical Generative Adversarial Networks (Graphical-GAN), which combines Bayesian networks for dependency structures and GANs for expressive functions, resulting in successful learning of discrete and temporal structures on visual datasets.
We propose Graphical Generative Adversarial Networks (Graphical-GAN) to model structured data. Graphical-GAN conjoins the power of Bayesian networks on compactly representing the dependency structures among random variables and that of generative adversarial networks on learning expressive dependency functions. We introduce a structured recognition model to infer the posterior distribution of latent variables given observations. We generalize the Expectation Propagation (EP) algorithm to learn the generative model and recognition model jointly. Finally, we present two important instances of Graphical-GAN, i.e. Gaussian Mixture GAN (GMGAN) and State Space GAN (SSGAN), which can successfully learn the discrete and temporal structures on visual datasets, respectively.