Semi-Discrete Normalizing Flows through Differentiable Tessellation
Abstract
Mapping between discrete and continuous distributions is a difficult task and many have had to resort to heuristical approaches. We propose a tessellation-based approach that directly learns quantization boundaries in a continuous space, complete with exact likelihood evaluations. This is done through constructing normalizing flows on convex polytopes parameterized using a simple homeomorphism with an efficient log determinant Jacobian. We explore this approach in two application settings, mapping from discrete to continuous and vice versa. Firstly, a Voronoi dequantization allows automatically learning quantization boundaries in a multidimensional space. The location of boundaries and distances between regions can encode useful structural relations between the quantized discrete values. Secondly, a Voronoi mixture model has near-constant computation cost for likelihood evaluation regardless of the number of mixture components. Empirically, we show improvements over existing methods across a range of structured data modalities.
Cite
@article{arxiv.2203.06832,
title = {Semi-Discrete Normalizing Flows through Differentiable Tessellation},
author = {Ricky T. Q. Chen and Brandon Amos and Maximilian Nickel},
journal= {arXiv preprint arXiv:2203.06832},
year = {2022}
}