Likelihood Contribution based Multi-scale Architecture for Generative Flows. (arXiv:1908.01686v3 [cs.LG] UPDATED)

Web: http://arxiv.org/abs/1908.01686

Deep generative modeling using flows has gained popularity owing to the
tractable exact log-likelihood estimation with efficient training and synthesis
process. However, flow models suffer from the challenge of having high
dimensional latent space, the same in dimension as the input space. An
effective solution to the above challenge as proposed by Dinh et al. (2016) is
a multi-scale architecture, which is based on iterative early factorization of
a part of the total dimensions at regular intervals. Prior works on …

architecture arxiv scale