Transformation Autoregressive Networks
[edit]
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:38983907, 2018.
Abstract
The fundamental task of general density estimation $p(x)$ has been of keen interest to machine learning. In this work, we attempt to systematically characterize methods for density estimation. Broadly speaking, most of the existing methods can be categorized into either using: a) autoregressive models to estimate the conditional factors of the chain rule, $p(x_{i}\, \, x_{i1}, \ldots)$; or b) nonlinear transformations of variables of a simple base distribution. Based on the study of the characteristics of these categories, we propose multiple novel methods for each category. For example we propose RNN based transformations to model nonMarkovian dependencies. Further, through a comprehensive study over both real world and synthetic data, we show that jointly leveraging transformations of variables and autoregressive conditional models, results in a considerable improvement in performance. We illustrate the use of our models in outlier detection and image modeling. Finally we introduce a novel data driven framework for learning a family of distributions.
Related Material


