DeepAI AI Chat
Log In Sign Up

Reparameterizing the Birkhoff Polytope for Variational Permutation Inference

by   Scott W. Linderman, et al.

Many matching, tracking, sorting, and ranking problems require probabilistic reasoning about possible permutations, a set that grows factorially with dimension. Combinatorial optimization algorithms may enable efficient point estimation, but fully Bayesian inference poses a severe challenge in this high-dimensional, discrete space. To surmount this challenge, we start with the usual step of relaxing a discrete set (here, of permutation matrices) to its convex hull, which here is the Birkhoff polytope: the set of all doubly-stochastic matrices. We then introduce two novel transformations: first, an invertible and differentiable stick-breaking procedure that maps unconstrained space to the Birkhoff polytope; second, a map that rounds points toward the vertices of the polytope. Both transformations include a temperature parameter that, in the limit, concentrates the densities on permutation matrices. We then exploit these transformations and reparameterization gradients to introduce variational inference over permutation matrices, and we demonstrate its utility in a series of experiments.


page 1

page 2

page 3

page 4


Cryptanalysis of quantum permutation pad

Cryptanalysis increases the level of confidence in cryptographic algorit...

Stochastic Optimization of Sorting Networks via Continuous Relaxations

Sorting input objects is an important step in many machine learning pipe...

Improved Variational Bayesian Phylogenetic Inference with Normalizing Flows

Variational Bayesian phylogenetic inference (VBPI) provides a promising ...

Learning Permutations with Sinkhorn Policy Gradient

Many problems at the intersection of combinatorics and computer science ...

General Bayesian Inference over the Stiefel Manifold via the Givens Transform

We introduce the Givens Transform, a novel transform between the space o...

DeepPermNet: Visual Permutation Learning

We present a principled approach to uncover the structure of visual data...

In Honour of Ted Swart

This is a tribute to my dear life-long friend, mentor and colleague Ted ...