English
Related papers

Related papers: Lost Relatives of the Gumbel Trick

200 papers

The Gumbel-max trick is a method to draw a sample from a categorical distribution, given by its unnormalized (log-)probabilities. Over the past years, the machine learning community has proposed several extensions of this trick to…

Machine Learning · Computer Science 2022-03-09 Iris A. M. Huijben , Wouter Kool , Max B. Paulus , Ruud J. G. van Sloun

The Gumbel-Max trick is the basis of many relaxed gradient estimators. These estimators are easy to implement and low variance, but the goal of scaling them comprehensively to large combinatorial distributions is still outstanding. Working…

Machine Learning · Statistics 2021-03-02 Max B. Paulus , Dami Choi , Daniel Tarlow , Andreas Krause , Chris J. Maddison

The Gumbel-Softmax probability distribution allows learning discrete tokens in generative learning, while the Gumbel-Argmax probability distribution is useful in learning discrete structures in discriminative learning. Despite the efforts…

Machine Learning · Computer Science 2024-06-05 Hedda Cohen Indelman , Tamir Hazan

The well-known Gumbel-Max Trick for sampling elements from a categorical distribution (or more generally a non-negative vector) and its variants have been widely used in areas such as machine learning and information retrieval. To sample a…

Machine Learning · Computer Science 2023-02-13 Yuanming Zhang , Pinghui Wang , Yiyan Qi , Kuankuan Cheng , Junzhou Zhao , Guangjian Tian , Xiaohong Guan

The well-known Gumbel-Max Trick for sampling elements from a categorical distribution (or more generally a nonnegative vector) and its variants have been widely used in areas such as machine learning and information retrieval. To sample a…

Computation · Statistics 2020-02-04 Yiyan Qi , Pinghui Wang , Yuanming Zhang , Junzhou Zhao , Guangjian Tian , Xiaohong Guan

A discrete version of the Gumbel (Type I) extreme value distribution has been derived by using the general approach of discretization of a continuous distribution. Important distributional and reliability properties have been explored. It…

Statistics Theory · Mathematics 2014-10-29 Subrata Chakraborty , Dhrubajyoti Chakravarty

The problem of drawing samples from a discrete distribution can be converted into a discrete optimization problem. In this work, we show how sampling from a continuous distribution can be converted into an optimization problem over…

Computation · Statistics 2015-01-27 Chris J. Maddison , Daniel Tarlow , Tom Minka

Assessing the significance of alignment scores of optimally aligned DNA or amino acid sequences can be achieved via the knowledge of the score distribution of random sequences. But this requires obtaining the distribution in the…

Quantitative Methods · Quantitative Biology 2016-08-24 Pascal Fieth , Alexander K. Hartmann

The gapped local alignment score of two random sequences follows a Gumbel distribution. If computers could estimate the parameters of the Gumbel distribution within one second, the use of arbitrary alignment scoring schemes could increase…

Statistics Theory · Mathematics 2009-09-04 Yonil Park , Sergey Sheetlin , John L. Spouge

Usual estimation methods for the parameters of extreme values distribution employ only a few values, wasting a lot of information. More precisely, in the case of the Gumbel distribution, only the block maxima values are used. In this work,…

Data Analysis, Statistics and Probability · Physics 2019-02-22 Rubén Gómez González , M. Isabel Parra , Francisco Javier Acero , Jacinto Martín

Many machine learning tasks require sampling a subset of items from a collection based on a parameterized distribution. The Gumbel-softmax trick can be used to sample a single item, and allows for low-variance reparameterized gradients with…

Machine Learning · Computer Science 2021-03-02 Sang Michael Xie , Stefano Ermon

A new unimodal distribution family indexed by the mode and three other parameters is derived from a mixture of a Gumbel distribution for the maximum and a Gumbel distribution for the minimum. Properties of the proposed distribution are…

Methodology · Statistics 2024-07-02 Qingyang Liu , Xianzheng Huang , Haiming Zhou

The particle Gibbs sampler is a Markov chain Monte Carlo (MCMC) algorithm to sample from the full posterior distribution of a state-space model. It does so by executing Gibbs sampling steps on an extended target distribution defined on the…

Computation · Statistics 2015-07-29 Nicolas Chopin , Sumeetpal S. Singh

We study a relaxation of the problem of coupling probability distributions -- a list of samples is generated from one distribution and an accept is declared if any one of these samples is identical to the sample generated from the other…

Machine Learning · Computer Science 2026-01-13 Joseph Rowan , Buu Phan , Ashish Khisti

Structured latent variables allow incorporating meaningful prior knowledge into deep learning models. However, learning with such variables remains challenging because of their discrete nature. Nowadays, the standard learning approach is to…

Machine Learning · Computer Science 2021-10-29 Kirill Struminsky , Artyom Gadetsky , Denis Rakitin , Danil Karpushkin , Dmitry Vetrov

In this work, we adopt a general framework based on the Gibbs posterior to update belief distributions for inverse problems governed by partial differential equations (PDEs). The Gibbs posterior formulation is a generalization of standard…

Computation · Statistics 2019-07-04 Zilong Zou , Sayan Mukherjee , Harbir Antil , Wilkins Aquino

We consider the problem of inference in discrete probabilistic models, that is, distributions over subsets of a finite ground set. These encompass a range of well-known models in machine learning, such as determinantal point processes and…

Machine Learning · Computer Science 2018-07-10 Alkis Gotovos , Hamed Hassani , Andreas Krause , Stefanie Jegelka

Estimating the gradients of stochastic nodes in stochastic computational graphs is one of the crucial research questions in the deep generative modeling community, which enables the gradient descent optimization on neural network…

Machine Learning · Computer Science 2023-02-23 Weonyoung Joo , Dongjun Kim , Seungjae Shin , Il-Chul Moon

A new method to simulate probability distributions in regions where the events are VERY unlikely (e.g. p ~ 10^{-40}) is presented. The basic idea is to represent the underlying probability space by the phase space of a physical system. The…

Disordered Systems and Neural Networks · Physics 2009-11-07 Alexander K. Hartmann

In this paper we describe how MAP inference can be used to sample efficiently from Gibbs distributions. Specifically, we provide means for drawing either approximate or unbiased samples from Gibbs' distributions by introducing low…

Machine Learning · Computer Science 2013-10-01 Tamir Hazan , Subhransu Maji , Tommi Jaakkola
‹ Prev 1 2 3 10 Next ›