Skip to yearly menu bar Skip to main content


Poster

Semi-Discrete Normalizing Flows through Differentiable Tessellation

Ricky T. Q. Chen · Brandon Amos · Maximilian Nickel

Hall J (level 1) #429

Keywords: [ dequantization ] [ disjoint mixture modeling ] [ deep probabilistic modeling ] [ normalizing flows ]


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.

Chat is not available.