Unsupervised Learning of Global Factors in Deep Generative Models

Ignacio Peis, Pablo M. Olmos, Antonio Artés

Keywords: unsupervised, autoencoders, disentanglement, generative models, representation learning

Abstract: We present a novel deep generative model based on non i.i.d. variational autoencoders that captures global dependencies among observations in a fully unsupervised fashion. In contrast to the recent semi-supervised alternatives for global modeling in deep generative models, our approach combines a mixture model in the local or data-dependent space and a global Gaussian latent variable, which lead us to obtain three particular insights. First, the induced latent global space captures interpretable disentangled representations with no user-defined regularization in the evidence lower bound (as in beta-VAE and its generalizations). Second, we show that the model performs domain alignment to find correlations and interpolate between different databases. Finally, we study the ability of the global space to discriminate between groups of observations with non-trivial underlying structures, such as face images with shared attributes or defined sequences of digits images.

One-sentence Summary: We present a deep generative model based on non i.i.d. VAEs for capturing global dependencies among observations in a fully unsupervised fashion, leading to global disentanglement, domain alignment and detection of underlying structures.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2012.08234/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=Qmkq6SE0HB