ICML 2019
Multivariate-Information Adversarial Ensemble for Scalable Joint Distribution Matching
Ziliang Chen, Zhanfu Yang, Xiaoxi Wang, Xiaodan Liang, Xiaopeng Yan, Guanbin Li, and Liang Lin*
ICML 2019


A broad range of cross-m-domain generation researches boil down to matching a joint distribution by deep generative models (DGMs). Hitherto algorithms excel in pairwise domains while as m 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 m-domain joint distribution matching. As an m-domain ensemble model of ALIs (Dumoulin et al., 2016), 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 m-domain joint distributions. MMI-ALI linearly scales as m increases and thus, strikes a right balance between efficacy and scalability. We evaluate MMI-ALI in diverse challenging m-domain scenarios and verify its superiority














In this paper, we have delved into the problem of multiple domain joint distribution matching that summarized a variety of cross-domain generation tasks. Instead of hacking a complex DGM pipeline, we propose MMI-ALI, which reshapes classical ALI from the perspective of model integration and is linearly-scalable with the domain number. It learns with an adversarial ensemble loss and can be applied in both supervised and unsupervised learning schemes. Extensive evaluation results on diverse m-domain scenarios have demonstrated the superiority of the proposed framework to the existing DGMs feasible for cross-m-domain generation, e.g., CycleGAN and Star-GAN.