We demonstrate how a variational approximation scheme enables effective inference of key parameters in probabilisitic signal models which employ the Student-t distribution. Using the two scenarios of previous termrobustnext term interpolation and independent component analysis (ICA) as examples, we illustrate the key feature of the approach: that the form of the noise distribution in the interpolation case, and the source distributions in the ICA case, can be inferred from the data concurrent with all other model parameters.

@Article{tipping-variational05,
title = {Variational inference for Student-$t$ models: Robust Bayesian interpolation and generalised component analysis},
journal = {Neurocomputing},
author = {Michael E. Tipping and Neil D. Lawrence},
pages = {123},
year = {2005},
volume = {69},
month = {00},
edit = {https://github.com/lawrennd//publications/edit/gh-pages/_posts/2005-01-01-tipping-variational05.md},
url = {http://inverseprobability.com/publications/tipping-variational05.html},
abstract = {We demonstrate how a variational approximation scheme enables effective inference of key parameters in probabilisitic signal models which employ the Student-t distribution. Using the two scenarios of previous termrobustnext term interpolation and independent component analysis (ICA) as examples, we illustrate the key feature of the approach: that the form of the noise distribution in the interpolation case, and the source distributions in the ICA case, can be inferred from the data concurrent with all other model parameters.},
key = {Tipping-variational05},
OPTgroup = {}
}

%T Variational inference for Student-$t$ models: Robust Bayesian interpolation and generalised component analysis
%A Michael E. Tipping and Neil D. Lawrence
%B
%C Neurocomputing
%D
%F tipping-variational05
%J Neurocomputing
%P 123--141
%R
%U http://inverseprobability.com/publications/tipping-variational05.html
%V 69
%X We demonstrate how a variational approximation scheme enables effective inference of key parameters in probabilisitic signal models which employ the Student-t distribution. Using the two scenarios of previous termrobustnext term interpolation and independent component analysis (ICA) as examples, we illustrate the key feature of the approach: that the form of the noise distribution in the interpolation case, and the source distributions in the ICA case, can be inferred from the data concurrent with all other model parameters.

TY - CPAPER
TI - Variational inference for Student-$t$ models: Robust Bayesian interpolation and generalised component analysis
AU - Michael E. Tipping
AU - Neil D. Lawrence
PY - 2005/01/01
DA - 2005/01/01
ID - tipping-variational05
SP - 123
EP - 141
UR - http://inverseprobability.com/publications/tipping-variational05.html
AB - We demonstrate how a variational approximation scheme enables effective inference of key parameters in probabilisitic signal models which employ the Student-t distribution. Using the two scenarios of previous termrobustnext term interpolation and independent component analysis (ICA) as examples, we illustrate the key feature of the approach: that the form of the noise distribution in the interpolation case, and the source distributions in the ICA case, can be inferred from the data concurrent with all other model parameters.
ER -

Tipping, M.E. & Lawrence, N.D.. (2005). Variational inference for Student-$t$ models: Robust Bayesian interpolation and generalised component analysis. Neurocomputing 69:123-141