The continuous categorical: a novel simplex-valued exponential family

by   Elliott Gordon-Rodriguez, et al.

Simplex-valued data appear throughout statistics and machine learning, for example in the context of transfer learning and compression of deep networks. Existing models for this class of data rely on the Dirichlet distribution or other related loss functions; here we show these standard choices suffer systematically from a number of limitations, including bias and numerical issues that frustrate the use of flexible network models upstream of these distributions. We resolve these limitations by introducing a novel exponential family of distributions for modeling simplex-valued data - the continuous categorical, which arises as a nontrivial multivariate generalization of the recently discovered continuous Bernoulli. Unlike the Dirichlet and other typical choices, the continuous categorical results in a well-behaved probabilistic loss function that produces unbiased estimators, while preserving the mathematical simplicity of the Dirichlet. As well as exploring its theoretical properties, we introduce sampling methods for this distribution that are amenable to the reparameterization trick, and evaluate their performance. Lastly, we demonstrate that the continuous categorical outperforms standard choices empirically, across a simulation study, an applied example on multi-party elections, and a neural network compression task.



There are no comments yet.


page 1

page 2

page 3

page 4


A conjugate prior for the Dirichlet distribution

This note investigates a conjugate class for the Dirichlet distribution ...

Fast Predictive Uncertainty for Classification with Bayesian Deep Networks

In Bayesian Deep Learning, distributions over the output of classificati...

Uses and Abuses of the Cross-Entropy Loss: Case Studies in Modern Deep Learning

Modern deep learning is primarily an experimental science, in which empi...

Gaussian-Dirichlet Posterior Dominance in Sequential Learning

We consider the problem of sequential learning from categorical observat...

Fast Maximum Likelihood Estimation and Supervised Classification for the Beta-Liouville Multinomial

The multinomial and related distributions have long been used to model c...

Morphology on categorical distributions

The categorical distribution is a natural representation of uncertainty ...

A Conway-Maxwell-Multinomial Distribution for Flexible Modeling of Clustered Categorical Data

Categorical data are often observed as counts resulting from a fixed num...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.