Learning Latent Permutations with Gumbel-Sinkhorn Networks

2018-02-23 · ICLR 2018 1 · Gonzalo Mena, David Belanger, Scott Linderman, Jasper Snoek

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 marginalization over these combinatorial objects is intractable. In response, this paper introduces a collection of new methods for end-to-end learning in such models that approximate discrete maximum-weight matching using the continuous Sinkhorn operator. Sinkhorn iteration is attractive because it functions as a simple, easy-to-implement analog of the softmax operator. With this, we can define the Gumbel-Sinkhorn method, an extension of the Gumbel-Softmax method (Jang et al. 2016, Maddison2016 et al. 2016) to distributions over latent matchings. We demonstrate the effectiveness of our method by outperforming competitive baselines on a range of qualitatively different tasks: sorting numbers, solving jigsaw puzzles, and identifying neural signals in worms.

📄 PDF Abstract BibTeX arXiv:1802.08665

Code (2)

google/gumbel_sinkhorn 공식 구현 tf

HeddaCohenIndelman/Learning-Gumbel-Sinkhorn-Permutations-w-Pytorch pytorch

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Foresight of Graph Reinforcement Learning Latent Permutations Learnt by Gumbel Sinkhorn Network

2021-10-23 · Tianqi Shen, Hong Zhang, Ding Yuan, Jiaping Xiao 외

Vital importance has necessity to be attached to cooperation in multi-agent environments, as a result of which some reinforcement learning algorithms combined with graph neural networks have been proposed to understand t…

Graph Attention reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Learning Permutation from Structure Without Supervision

2026-05-25 · Ran Eisenberg, Ofir Lindenbaum arxiv

Many learning problems require uncovering a hidden ordering that reveals structure in unordered data, such as monotonicity in sorting or spatial continuity in jigsaw reconstruction. In these settings, permutations can be…

Sparse Sinkhorn Attention

2020-02-26 · ICML 2020 1 · Yi Tay, Dara Bahri, Liu Yang, Donald Metzler 외

We propose Sparse Sinkhorn Attention, a new efficient and sparse method for learning to attend. Our method is based on differentiable sorting of internal representations. Concretely, we introduce a meta sorting network t…

Document Classification Image Generation Language Modeling Language Modelling+1

Learning Permutations with Sinkhorn Policy Gradient

2018-05-18 · Patrick Emami, Sanjay Ranka

Many problems at the intersection of combinatorics and computer science require solving for a permutation that optimally matches, ranks, or sorts some data. These problems usually have a task-specific, often non-differen…

Representation Learning

Learning Unbiased Permutations via Flow Matching

2026-05-16 · Yimeng Min, Carla P. Gomes arxiv

Learning permutations is fundamental to sorting, ranking, and matching, but existing differentiable methods based on entropy-regularized Sinkhorn produce a single softened solution and collapse under ambiguity. We presen…