Gaussian, Mean Field and Variational Approximation: the Equivalence
نویسنده
چکیده
We start by clarifying the terms which appear in the title. For a given theory of selfinteracting quantum fields, by Gaussian approximation we mean the Gaussian part of the interacting measure. By mean field approximation we understand the leading term of the expansion given in [1, 2] and by variational approximation we understand the old variational technique performed with some particular trial states [3-5]. The novelty of this paper is the method of extracting the Gaussian peace of the interacting measure. Also we hope that the equivalence between the three methods will lead to a better understanding of the self-interacting field theories. The expansion around the mean field (MF) approximation have been successfully used in [1, 2] to λ ∑N n=1 αn (λφ) 2n interactions, for small λ. The key role of this approximation is that, after the translation to φ− φMF , it brings all the coupling constants near to zero where the cluster expansion [6] can be applied.
منابع مشابه
Correlated Gaussian method for dilute bosonic systems
Abstract. The weakly interacting trapped Bose gases have been customarily described using the mean-field approximation in the form of the Gross-Pitaevskii equation. The mean-field approximation, however, has certain limitations, in particular it can not describe correlations between particles. We introduce here an alternative variational approach, based on the correlated Gaussian method, which ...
متن کاملVariational mean-field algorithm for efficient inference in large systems of stochastic differential equations.
This work introduces a Gaussian variational mean-field approximation for inference in dynamical systems which can be modeled by ordinary stochastic differential equations. This new approach allows one to express the variational free energy as a functional of the marginal moments of the approximating Gaussian process. A restriction of the moment equations to piecewise polynomial functions, over ...
متن کاملBeta Process Non-negative Matrix Factorization with Stochastic Structured Mean-Field Variational Inference
Beta process is the standard nonparametric Bayesian prior for latent factor model. In this paper, we derive a structured mean-field variational inference algorithm for a beta process non-negative matrix factorization (NMF) model with Poisson likelihood. Unlike the linear Gaussian model, which is well-studied in the nonparametric Bayesian literature, NMF model with beta process prior does not en...
متن کاملPolya-gamma augmentations for factor models
Bayesian inference for latent factor models, such as principal component and canonical correlation analysis, is easy for Gaussian likelihoods with conjugate priors using both Gibbs sampling and mean-field variational approximation. For other likelihood potentials one needs to either resort to more complex sampling schemes or to specifying dedicated forms for variational lower bounds. Recently, ...
متن کاملVariational Multinomial Logit Gaussian Process
Gaussian process prior with an appropriate likelihood function is a flexible non-parametric model for a variety of learning tasks. One important and standard task is multi-class classification, which is the categorization of an item into one of several fixed classes. A usual likelihood function for this is the multinomial logistic likelihood function. However, exact inference with this model ha...
متن کامل