Open AccessProceedings Article
Variational Linear Response
Manfred Opper,Ole Winther +1 more
- 09 Dec 2003
- Vol. 16, pp 1157-1164
TL;DR: A general linear response method for deriving improved estimates of correlations in the variational Bayes framework is presented and it is discussed how to use linear response as a general principle for improving mean field approximations.
read more
Abstract: A general linear response method for deriving improved estimates of correlations in the variational Bayes framework is presented. Three applications are given and it is discussed how to use linear response as a general principle for improving mean field approximations.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
•Journal Article
Expectation Consistent Approximate Inference
Manfred Opper,Ole Winther +1 more
TL;DR: A novel framework for approximations to intractable probabilistic models which is based on a free energy formulation is proposed which requires two tractable probability distributions which are made consistent on a set of moments and encode different features of the originalintractable distribution.
•Posted Content
Covariances, Robustness, and Variational Bayes
TL;DR: This paper derived a simple formula for the effect of infinitesimal model perturbations on MFVB posterior means and provided improved covariance estimates and local robustness measures for MFVB.
79
•Proceedings Article
Linear response methods for accurate covariance estimates from Mean field variational Bayes
Ryan Giordano,Tamara Broderick,Michael I. Jordan +2 more
- 07 Dec 2015
TL;DR: The linear response variational Bayes (LRVB) as mentioned in this paper generalizes linear response methods from statistical physics to deliver accurate uncertainty estimates for model variables, both for individual variables and coherently across variables.
Bethe–Peierls approximation and the inverse Ising problem
H. Chau Nguyen,Johannes Berg +1 more
TL;DR: The authors applied the Bethe-Peierls approximation to the inverse Ising problem and showed how the linear response relation leads to a simple method for reconstructing couplings and fields of the Ising model.
58
•Posted Content
Linear Response Methods for Accurate Covariance Estimates from Mean Field Variational Bayes
TL;DR: This work generalizes linear response methods from statistical physics to deliver accurate uncertainty estimates for model variables—both for individual variables and coherently across variables.
52
References
•Book
Information Theory, Inference and Learning Algorithms
David J. C. MacKay
- 06 Oct 2003
TL;DR: A fun and exciting textbook on the mathematics underpinning the most dynamic areas of modern science and engineering.
•Book
Statistical Field Theory
Giorgio Parisi,Jonathan Machta +1 more
- 01 Jan 1988
TL;DR: In this article, the authors present a model for classical equilibrium statistical mechanics, including the Ising model, the Landau-Ginsbourg model, and the Renormalization Group.
1K
•Proceedings Article
A Variational Baysian Framework for Graphical Models
Hagai Attias
- 29 Nov 1999
TL;DR: This paper presents a novel practical framework for Bayesian model averaging and model selection in probabilistic graphical models that approximates full posterior distributions over model parameters and structures, as well as latent variables, in an analytical manner.
Mean-field approaches to independent component analysis
TL;DR: Three increasingly advanced mean-field methods are investigated: the variational (also known as naive mean field) approach, linear response corrections, and an adaptive version of the Thouless, Anderson and Palmer (1977) (TAP) mean- field approach, which is due to Opper and Winther (2001).
•Proceedings Article
Structured Variational Distributions in VIBES
Christopher M. Bishop,John Winn +1 more
- 03 Jan 2003
TL;DR: This paper presents an extension of VIBES in which the variational posterior distribution corresponds to a sub-graph of the full probabilistic model, which can produce much closer approximations to the true posterior distribution.