Gaussian Kullback-Leibler approximate inference

By Edward Challis, David Barber. 2013

78 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Variational Inference: A Review for Statisticians (2016, neutral): “…recently, Hoffman and Blei (2015) used generic structured variational inference in a stochastic optimization algorithm; Kucukelbir et al. (2017), Challis and Barber (2013), and Tan and Nott (2017) took advantage of Gaussian variational families with nondiagonal covariance; Giordano, Broderick,…”
Variational Dropout Sparsifies Deep Neural Networks (2017, background): “Recently several works on efﬁcient training of Sparse Bayesian Models have appeared (Challis & Barber, 2013; Titsias & L´azaro-Gredilla, 2014).”
Automatic Differentiation Variational Inference (2016, background): “Various low-rank approximations to the covariance matrix reduce this cost, yet limit its ability to model complex posterior correlations (Seeger, 2010; Challis and Barber, 2013).”
Neural Importance Sampling (2018, neutral): “iational methods [Jordan et al. 1999] consider approximating the true posterior distribution with another distribution q(z|x;ϕ), which is chosen to have a simple functional form: commonly a Gaussian [Challis and Barber 2013] or a distribution which factorizes across variables [Ghahramani et al. 2000”
Gaussian variational approximation with sparse precision matrices (2016, background): “The relationship between the Laplace and the Gaussian variational approximation is discussed in Opper and Archambeau (2009) while Challis and Barber (2013) consider some different parameterizations in terms of the Cholesky.”
Semiparametric Mean Field Variational Bayes: General Principles and Numerical Issues (2016, methodology): “This implies the possibility of low-rank approximations to the above semiparametric mean field variational Bayes algorithm given, as described in Section 4.1.3 of Challis and Barber (2013).”
Doubly Stochastic Variational Bayes for non-Conjugate Inference (2014, methodology): “This generalises the result of Challis & Barber (2011; 2013), who proved it for the variational Gaussian approximation, and it is similar to the generalisation presented in (Staines & Barber, 2012).”
Variational Inference for Heteroscedastic Semiparametric Regression (2015, methodology): “…passing that we use involves approximating particular posterior density functions by Multivariate Normal density functions, which is an established practice within the variational approximation literature (e.g. Hinton & van Camp 1993; Barber & Bishop 1997; Raiko et al. 2007; Challis & Barber 2013).”