JointGAN: Multi-Domain Joint Distribution Learning with Generative Adversarial Nets
This work addresses a fundamental limitation in generative modeling for multi-domain data, though it appears incremental as an extension of GANs to joint distribution learning.
The authors tackled the problem of learning joint distributions across multiple domains, which existing GANs typically do not address, by developing JointGAN, a framework that synthesizes samples from marginals, conditionals, and the full joint distribution using multiple generators and a single critic.
A new generative adversarial network is developed for joint distribution matching. Distinct from most existing approaches, that only learn conditional distributions, the proposed model aims to learn a joint distribution of multiple random variables (domains). This is achieved by learning to sample from conditional distributions between the domains, while simultaneously learning to sample from the marginals of each individual domain. The proposed framework consists of multiple generators and a single softmax-based critic, all jointly trained via adversarial learning. From a simple noise source, the proposed framework allows synthesis of draws from the marginals, conditional draws given observations from a subset of random variables, or complete draws from the full joint distribution. Most examples considered are for joint analysis of two domains, with examples for three domains also presented.