English
Related papers

Related papers: Sticking the Landing: Simple, Lower-Variance Gradi…

200 papers

Stochastic neurons can be useful for a number of reasons in deep learning models, but in many cases they pose a challenging problem: how to estimate the gradient of a loss function with respect to the input of such stochastic neurons, i.e.,…

Machine Learning · Computer Science 2013-05-15 Yoshua Bengio

Gradient-based methods for optimisation of objectives in stochastic settings with unknown or intractable dynamics require estimators of derivatives. We derive an objective that, under automatic differentiation, produces low-variance…

Machine Learning · Computer Science 2019-09-25 Gregory Farquhar , Shimon Whiteson , Jakob Foerster

Recent progress in deep latent variable models has largely been driven by the development of flexible and scalable variational inference methods. Variational training of this type involves maximizing a lower bound on the log-likelihood,…

Machine Learning · Computer Science 2016-06-02 Andriy Mnih , Danilo J. Rezende

We study the performance of estimators of a sparse nonrandom vector based on an observation which is linearly transformed and corrupted by additive white Gaussian noise. Using the reproducing kernel Hilbert space framework, we derive a new…

Statistics Theory · Mathematics 2010-09-20 Sebastian Schmutzhard , Alexander Jung , Franz Hlawatsch , Zvika Ben-Haim , Yonina C. Eldar

We deal with the problem of gradient estimation for stochastic differentiable relaxations of algorithms, operators, simulators, and other non-differentiable functions. Stochastic smoothing conventionally perturbs the input of a…

Machine Learning · Computer Science 2024-10-11 Felix Petersen , Christian Borgelt , Aashwin Mishra , Stefano Ermon

Models with a large number of latent variables are often used to fully utilize the information in big or complex data. However, they can be difficult to estimate using standard approaches, and variational inference methods are a popular…

Methodology · Statistics 2021-04-20 Rubén Loaiza-Maya , Michael Stanley Smith , David J. Nott , Peter J. Danaher

The learning and evaluation of energy-based latent variable models (EBLVMs) without any structural assumptions are highly challenging, because the true posteriors and the partition functions in such models are generally intractable. This…

Machine Learning · Computer Science 2021-06-08 Fan Bao , Kun Xu , Chongxuan Li , Lanqing Hong , Jun Zhu , Bo Zhang

This paper presents a practical and simple fully nonparametric multivariate smoothing procedure that adapts to the underlying smoothness of the true regression function. Our estimator is easily computed by successive application of existing…

Methodology · Statistics 2011-06-08 P. A. Cornillon , N. Hengartner , E. Matzner-Løber

Stochastic gradient-based optimisation for discrete latent variable models is challenging due to the high variance of gradients. We introduce a variance reduction technique for score function estimators that makes use of double control…

Machine Learning · Statistics 2022-06-07 Michalis K. Titsias , Jiaxin Shi

Deep latent variable models have become a popular model choice due to the scalable learning algorithms introduced by (Kingma & Welling, 2013; Rezende et al., 2014). These approaches maximize a variational lower bound on the intractable log…

Machine Learning · Computer Science 2018-11-20 George Tucker , Dieterich Lawson , Shixiang Gu , Chris J. Maddison

We compute a variance lower bound for unbiased estimators in specified statistical models. The construction of the bound is related to the original Cram\'er-Rao bound, although it does not require the differentiability of the model.…

Statistics Theory · Mathematics 2012-04-13 Thibault Espinasse , Paul Rochet

In this paper, we study the estimation of the derivative of a regression function in a standard univariate regression model. The estimators are defined either by derivating nonparametric least-squares estimators of the regression function…

Statistics Theory · Mathematics 2023-11-13 Fabienne Comte , Nicolas Marie

We introduce a novel one-parameter variational objective that lower bounds the data evidence and enables the estimation of approximate fractional posteriors. We extend this framework to hierarchical construction and Bayes posteriors,…

Machine Learning · Computer Science 2026-03-31 Kian Ming A. Chai , Edwin V. Bonilla

The Black Box Variational Inference (Ranganath et al. (2014)) algorithm provides a universal method for Variational Inference, but taking advantage of special properties of the approximation family or of the target can improve the…

Computation · Statistics 2019-06-18 Alexander Immer , Guillaume P. Dehaene

Most zeroth-order optimization algorithms mimic a first-order algorithm but replace the gradient of the objective function with some gradient estimator that can be computed from a small number of function evaluations. This estimator is…

Optimization and Control · Mathematics 2026-01-12 Wouter Jongeneel , Man-Chung Yue , Daniel Kuhn

We rewrite the standard nodal virtual element method as a generalised gradient method. This re-formulation allows for computing a reliable and efficient error estimator by locally reconstructing broken fluxes and potentials on elemental…

Numerical Analysis · Mathematics 2025-03-18 Théophile Chaumont-Frelet , Joscha Gedicke , Lorenzo Mascotto

Variational inference is a powerful tool for approximate inference. However, it mainly focuses on the evidence lower bound as variational objective and the development of other measures for variational inference is a promising area of…

Machine Learning · Computer Science 2016-12-06 Michael Figurnov , Kirill Struminsky , Dmitry Vetrov

We propose an algorithm to estimate the path-gradient of both the reverse and forward Kullback-Leibler divergence for an arbitrary manifestly invertible normalizing flow. The resulting path-gradient estimators are straightforward to…

Machine Learning · Computer Science 2022-07-19 Lorenz Vaitl , Kim A. Nicoli , Shinichi Nakajima , Pan Kessel

We show a connection between the Fourier spectrum of Boolean functions and the REINFORCE gradient estimator for binary latent variable models. We show that REINFORCE estimates (up to a factor) the degree-1 Fourier coefficients of a Boolean…

Machine Learning · Computer Science 2018-08-14 Adeel Pervez

A novel switching differentiator that has considerably simple form is proposed. Under the assumption that time-derivatives of the signal are norm-bounded, it is shown that estimation errors are convergent to the zeros asymptotically. The…

Systems and Control · Computer Science 2018-05-01 Jang-Hyun Park