Learning Latent Partial Matchings with Gumbel-IPF Networks
Learning to match discrete objects has been a central task in machine learning, often facilitated by a continuous relaxation of the matching structure. However, practical problems entail partial matchings due to missing correspondences, which pose difficulties for the one-to-one matching learning techniques that dominate the state-of-the-art. This paper introduces Gumbel-IPF networks for learning latent partial matchings. At the core of our method is the differentiable Iterative Proportional Fitting (IPF) procedure that disproportionally projects onto the transportation polytope of target marginals. Our theoretical framework also allows drawing samples from the temperature-dependent partial matching distribution. We investigate the properties of common-practice relaxations through the lens of biproportional fitting and introduce a new metric, the empirical prediction shift. Our method’s advantages are demonstrated in experimental results on the semantic keypoints partial matching task on the Pascal VOC, IMC-PT-SparseGM, and CUB2001 datasets.
Code (1)
Tasks
Graph MatchingSimilar Papers 제목 키워드 기반
Learning Latent Permutations with Gumbel-Sinkhorn Networks
Permutations and matchings are core building blocks in a variety of latent variable models, as they allow us to align, canonicalize, and sort data. Learning in such models is difficult, however, because exact marginaliza…
Priority-Neutral Matching Lattices Are Not Distributive
Stable matchings are a cornerstone of market design, with numerous practical deployments backed by a rich, theoretically-tractable structure. However, in school-choice problems, stable matchings are not Pareto optimal fo…
On extensions of partial priorities in school choice
We consider a school choice matching model where the priorities for schools are represented by binary relations that may not be weak order. We focus on the (total order) extensions of the binary relations. We introduce a…
Cross Species Expression Analysis using a Dirichlet Process Mixture Model with Latent Matchings
Recent studies compare gene expression data across species to identify core and species specific genes in biological systems. To perform such comparisons researchers need to match genes across species. This is a challeng…
Synchronisation of Partial Multi-Matchings via Non-negative Factorisations
In this work we study permutation synchronisation for the challenging case of partial permutations, which plays an important role for the problem of matching multiple objects (e.g. images or shapes). The term synchronisa…
Clustering