中文
相关论文

相关论文: The Concrete Distribution: A Continuous Relaxation…

200 篇论文

Low-variance gradient estimation is crucial for learning directed graphical models parameterized by neural networks, where the reparameterization trick is widely used for those with continuous variables. While this technique gives…

机器学习 · 统计学 2016-11-07 Seiya Tokui , Issei sato

Variational inference using the reparameterization trick has enabled large-scale approximate Bayesian inference in complex probabilistic models, leveraging stochastic optimization to sidestep intractable expectations. The reparameterization…

机器学习 · 统计学 2020-02-13 Christian A. Naesseth , Francisco J. R. Ruiz , Scott W. Linderman , David M. Blei

The ability to backpropagate stochastic gradients through continuous latent distributions has been crucial to the emergence of variational autoencoders and stochastic gradient variational Bayes. The key ingredient is an unbiased and…

神经与进化计算 · 计算机科学 2016-07-20 Alex Graves

Learning models with categorical variables requires optimizing expectations over discrete distributions, a setting in which stochastic gradient-based optimization is challenging due to the non-differentiability of categorical sampling. A…

机器学习 · 计算机科学 2026-02-10 Samson Gourevitch , Alain Durmus , Eric Moulines , Jimmy Olsson , Yazid Janati

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…

机器学习 · 计算机科学 2023-02-23 Weonyoung Joo , Dongjun Kim , Seungjae Shin , Il-Chul Moon

We present a new algorithm for stochastic variational inference that targets at models with non-differentiable densities. One of the key challenges in stochastic variational inference is to come up with a low-variance estimator of the…

机器学习 · 计算机科学 2018-10-26 Wonyeol Lee , Hangyeol Yu , Hongseok Yang

By providing a simple and efficient way of computing low-variance gradients of continuous random variables, the reparameterization trick has become the technique of choice for training a variety of latent variable models. However, it is not…

机器学习 · 计算机科学 2019-01-31 Michael Figurnov , Shakir Mohamed , Andriy Mnih

Reparameterizable densities are an important way to learn probability distributions in a deep learning setting. For many distributions it is possible to create low-variance gradient estimators by utilizing a `reparameterization trick'. Due…

机器学习 · 统计学 2019-03-13 Luca Falorsi , Pim de Haan , Tim R. Davidson , Patrick Forré

Numerous models for supervised and reinforcement learning benefit from combinations of discrete and continuous model components. End-to-end learnable discrete-continuous models are compositional, tend to generalize better, and are more…

机器学习 · 计算机科学 2023-07-27 David Friede , Mathias Niepert

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…

机器学习 · 计算机科学 2021-02-09 Tomasz Kuśmierczyk , Arto Klami

The reparameterization gradient has become a widely used method to obtain Monte Carlo gradients to optimize the variational objective. However, this technique does not easily apply to commonly used distributions such as beta or gamma…

机器学习 · 统计学 2016-10-20 Francisco J. R. Ruiz , Michalis K. Titsias , David M. Blei

We propose a framework that lifts the capabilities of graph convolutional networks (GCNs) to scenarios where no input graph is given and increases their robustness to adversarial attacks. We formulate a joint probabilistic model that…

机器学习 · 计算机科学 2020-10-23 Pantelis Elinas , Edwin V. Bonilla , Louis Tiao

The reparameterization trick is widely used in variational inference as it yields more accurate estimates of the gradient of the variational objective than alternative approaches such as the score function method. Although there is…

机器学习 · 统计学 2018-12-31 Ming Xu , Matias Quiroz , Robert Kohn , Scott A. Sisson

We study the foundations of variational inference, which frames posterior inference as an optimisation problem, for probabilistic programming. The dominant approach for optimisation in practice is stochastic gradient descent. In particular,…

编程语言 · 计算机科学 2023-01-10 Basim Khajwal , C. -H. Luke Ong , Dominik Wagner

Gradient estimation -- approximating the gradient of an expectation with respect to the parameters of a distribution -- is central to the solution of many machine learning problems. However, when the distribution is discrete, most common…

机器学习 · 统计学 2024-04-16 Jiaxin Shi , Yuhao Zhou , Jessica Hwang , Michalis K. Titsias , Lester Mackey

The reparameterization trick has become one of the most useful tools in the field of variational inference. However, the reparameterization trick is based on the standardization transformation which restricts the scope of application of…

机器学习 · 计算机科学 2020-02-26 Anbang Wu , Shuangxi Chen , Chunming Wu

Learning in models with discrete latent variables is challenging due to high variance gradient estimators. Generally, approaches have relied on control variates to reduce the variance of the REINFORCE estimator. Recent work (Jang et al.…

机器学习 · 计算机科学 2017-11-07 George Tucker , Andriy Mnih , Chris J. Maddison , Dieterich Lawson , Jascha Sohl-Dickstein

We study the implicit regularization of gradient descent towards structured sparsity via a novel neural reparameterization, which we call a diagonally grouped linear neural network. We show the following intriguing property of our…

机器学习 · 统计学 2023-01-31 Jiangyuan Li , Thanh V. Nguyen , Chinmay Hegde , Raymond K. W. Wong

Training neural network models with discrete (categorical or structured) latent variables can be computationally challenging, due to the need for marginalization over large or combinatorial sets. To circumvent this issue, one typically…

机器学习 · 计算机科学 2020-12-29 Gonçalo M. Correia , Vlad Niculae , Wilker Aziz , André F. T. Martins

It is well-known that the reparameterisation gradient estimator, which exhibits low variance in practice, is biased for non-differentiable models. This may compromise correctness of gradient-based optimisation methods such as stochastic…

机器学习 · 计算机科学 2024-02-21 Dominik Wagner , Basim Khajwal , C. -H. Luke Ong
‹ 上一页 1 2 3 10 下一页 ›