Tensor Belief Propagation
Andrew Wrigley, Wee Sun Lee, Nan Ye
Abstract
We propose a new approximate inference algorithm for graphical models, tensor belief propagation, based on approximating the messages passed in the junction tree algorithm. Our algorithm represents the potential functions of the graphical model and all messages on the junction tree compactly as mixtures of rank-1 tensors. Using this representation, we show how to perform the operations required for inference on the junction tree efficiently: marginalisation can be computed quickly due to the factored form of rank-1 tensors while multiplication can be approximated using sampling. Our analysis gives sufficient conditions for the algorithm to perform well, including for the case of high-treewidth graphs, for which exact inference is intractable. We compare our algorithm experimentally with several approximate inference algorithms and show that it performs well.
BibTeX
@InProceedings{pmlr-v70-wrigley17a,
title = {Tensor Belief Propagation},
author = {Andrew Wrigley and Wee Sun Lee and Nan Ye},
booktitle = {Proceedings of the 34th International Conference on Machine Learning},
pages = {3771--3779},
year = {2017},
editor = {Precup, Doina and Teh, Yee Whye},
volume = {70},
series = {Proceedings of Machine Learning Research},
month = {06--11 Aug},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v70/wrigley17a/wrigley17a.pdf},
url = {https://proceedings.mlr.press/v70/wrigley17a.html},
abstract = {We propose a new approximate inference algorithm for graphical models, tensor belief propagation, based on approximating the messages passed in the junction tree algorithm. Our algorithm represents the potential functions of the graphical model and all messages on the junction tree compactly as mixtures of rank-1 tensors. Using this representation, we show how to perform the operations required for inference on the junction tree efficiently: marginalisation can be computed quickly due to the factored form of rank-1 tensors while multiplication can be approximated using sampling. Our analysis gives sufficient conditions for the algorithm to perform well, including for the case of high-treewidth graphs, for which exact inference is intractable. We compare our algorithm experimentally with several approximate inference algorithms and show that it performs well.}
}