TAP Gibbs Free Energy, Belief Propagation and Sparsity
Abstract
The adaptive TAP Gibbs free energy for a general densely connected probabilistic model with quadratic interactions and arbritary single site constraints is derived. We show how a specific sequential minimization of the free energy leads to a generalization of Minka’s expectation propa- gation. Lastly, we derive a sparse representation version of the sequential algorithm. The usefulness of the approach is demonstrated on classifica- tion and density estimation with Gaussian processes and on an indepen- dent component analysis problem.
Cite
Text
Csató et al. "TAP Gibbs Free Energy, Belief Propagation and Sparsity." Neural Information Processing Systems, 2001.Markdown
[Csató et al. "TAP Gibbs Free Energy, Belief Propagation and Sparsity." Neural Information Processing Systems, 2001.](https://mlanthology.org/neurips/2001/csato2001neurips-tap/)BibTeX
@inproceedings{csato2001neurips-tap,
title = {{TAP Gibbs Free Energy, Belief Propagation and Sparsity}},
author = {Csató, Lehel and Opper, Manfred and Winther, Ole},
booktitle = {Neural Information Processing Systems},
year = {2001},
pages = {657-663},
url = {https://mlanthology.org/neurips/2001/csato2001neurips-tap/}
}