English
Related papers

Related papers: Rao-Blackwellizing the Straight-Through Gumbel-Sof…

200 papers

In many applications we seek to maximize an expectation with respect to a distribution over discrete variables. Estimating gradients of such objectives with respect to the distribution parameters is a challenging problem. We analyze…

Machine Learning · Statistics 2019-06-18 Evgeny Andriyash , Arash Vahdat , Bill Macready

While deep generative models have succeeded in image processing, natural language processing, and reinforcement learning, training that involves discrete random variables remains challenging due to the high variance of its gradient…

Machine Learning · Computer Science 2022-06-16 Ting-Han Fan , Ta-Chung Chi , Alexander I. Rudnicky , Peter J. Ramadge

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

Latent Gaussian variables have been popularised in probabilistic machine learning. In turn, gradient estimators are the machinery that facilitates gradient-based optimisation for models with latent Gaussian variables. The reparameterisation…

Machine Learning · Statistics 2025-10-21 Kevin H. Lam , Thang D. Bui , George Deligiannidis , Yee Whye Teh

Machine learning models involving discrete latent variables require gradient estimators to facilitate backpropagation in a computationally efficient manner. The most recent addition to the Straight-Through family of estimators, ReinMax, can…

Machine Learning · Statistics 2026-03-10 Daniel Wang , Thang D. Bui

We derive an unbiased estimator for expectations over discrete random variables based on sampling without replacement, which reduces variance as it avoids duplicate samples. We show that our estimator can be derived as the…

Machine Learning · Computer Science 2020-02-17 Wouter Kool , Herke van Hoof , Max Welling

We wish to compute the gradient of an expectation over a finite or countably infinite sample space having $K \leq \infty$ categories. When $K$ is indeed infinite, or finite but very large, the relevant summation is intractable. Accordingly,…

Machine Learning · Statistics 2019-05-14 Runjing Liu , Jeffrey Regier , Nilesh Tripuraneni , Michael I. Jordan , Jon McAuliffe

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

Discrete and especially binary random variables occur in many machine learning models, notably in variational autoencoders with binary latent states and in stochastic binary networks. When learning such models, a key tool is an estimator of…

Machine Learning · Computer Science 2021-10-18 Alexander Shekhovtsov

Categorical variables are a natural choice for representing discrete structure in the world. However, stochastic neural networks rarely use categorical latent variables due to the inability to backpropagate through samples. In this work, we…

Machine Learning · Statistics 2017-08-08 Eric Jang , Shixiang Gu , Ben Poole

Reparameterization of variational auto-encoders with continuous random variables is an effective method for reducing the variance of their gradient estimates. In the discrete case, one can perform reparametrization using the Gumbel-Max…

Machine Learning · Computer Science 2019-12-10 Guy Lorberbom , Andreea Gane , Tommi Jaakkola , Tamir Hazan

Estimating the Kullback--Leibler (KL) divergence between language models has many applications, e.g., reinforcement learning from human feedback (RLHF), interpretability, and knowledge distillation. However, computing the exact KL…

Computation and Language · Computer Science 2025-10-28 Afra Amini , Tim Vieira , Ryan Cotterell

Training models with discrete latent variables is challenging due to the high variance of unbiased gradient estimators. While low-variance reparameterization gradients of a continuous relaxation can provide an effective solution, a…

Machine Learning · Computer Science 2022-11-16 Zhe Dong , Andriy Mnih , George Tucker

The Gumbel-Softmax is a continuous distribution over the simplex that is often used as a relaxation of discrete distributions. Because it can be readily interpreted and easily reparameterized, it enjoys widespread use. We propose a modular…

Machine Learning · Statistics 2022-08-30 Andres Potapczynski , Gabriel Loaiza-Ganem , John P. Cunningham

In domains such as molecular and protein generation, physical systems exhibit inherent symmetries that are critical to model. Two main strategies have emerged for learning invariant distributions: designing equivariant network architectures…

Machine Learning · Computer Science 2025-10-21 Vinh Tong , Hoang Trung-Dung , Anji Liu , Guy Van den Broeck , Mathias Niepert

Much research has been carried out on shrinkage methods for real-valued covariance matrices. In spectral analysis of $p$-vector-valued time series there is often a need for good shrinkage methods too, most notably when the complex-valued…

Statistics Theory · Mathematics 2015-10-28 A. T. Walden , D. Schneider-Luftman

Variational approximations are increasingly based on gradient-based optimization of expectations estimated by sampling. Handling discrete latent variables is then challenging because the sampling process is not differentiable. Continuous…

Machine Learning · Computer Science 2021-02-09 Tomasz Kuśmierczyk , Arto Klami

We study control variate methods for Markov chain Monte Carlo (MCMC) in the setting of deterministic sweep sampling using $K\geq 2$ transition kernels. New variance reduction results are provided for MCMC averages based on sweeps over…

Statistics Theory · Mathematics 2019-12-17 Stephen Berg , Jun Zhu , Murray K. Clayton

We address covariance estimation in the sense of minimum mean-squared error (MMSE) for Gaussian samples. Specifically, we consider shrinkage methods which are suitable for high dimensional problems with a small number of samples (large p…

Methodology · Statistics 2015-05-13 Yilun Chen , Ami Wiesel , Yonina C. Eldar , Alfred O. Hero

Reinforcement learning from human feedback (RLHF) has become the dominant paradigm for aligning large language models with human preferences. However, policy gradient methods such as PPO suffer from high variance gradient estimates,…

Machine Learning · Computer Science 2026-01-21 Lukas Abrie Nel
‹ Prev 1 2 3 10 Next ›