Multivariate-Information Adversarial Ensemble for Scalable Joint Distribution Matching

A broad range of cross--domain generation researches boil down to matching a joint distribution by deep generative models (DGMs). Hitherto algorithms excel in pairwise domains while as increases, remain struggling to scale themselves to fit a joint distribution. In this paper, we propose a domain-scalable DGM, i.e., MMI-ALI for -domain joint distribution matching. As an -domain ensemble model of ALIs \cite{dumoulin2016adversarially}, MMI-ALI is adversarially trained with maximizing Multivariate Mutual Information (MMI) w.r.t. joint variables of each pair of domains and their shared feature. The negative MMIs are upper bounded by a series of feasible losses that provably lead to matching -domain joint distributions. MMI-ALI linearly scales as increases and thus, strikes a right balance between efficacy and scalability. We evaluate MMI-ALI in diverse challenging -domain scenarios and verify its superiority.
View on arXiv