Lifted First-Order Probabilistic Inference
Abstract
Most probabilistic inference algorithms are specified and processed on a propositional level. In the last decade, many proposals for algorithms accepting first-order specifications have been presented, but in the inference stage they still operate on a mostly propositional representation level. [Poole, 2003] presented a method to perform inference directly on the first-order level, but this method is limited to special cases. In this paperwe present the first exact inference algorithm that operates directly on a first-order level, and that can be applied to any first-order model (specified in a language that generalizes undirected graphical models). Our experiments show superior performance in comparison with propositional exact inference.
Cite
Text
de Salvo Braz et al. "Lifted First-Order Probabilistic Inference." International Joint Conference on Artificial Intelligence, 2005. doi:10.7551/mitpress/7432.003.0017Markdown
[de Salvo Braz et al. "Lifted First-Order Probabilistic Inference." International Joint Conference on Artificial Intelligence, 2005.](https://mlanthology.org/ijcai/2005/desalvobraz2005ijcai-lifted/) doi:10.7551/mitpress/7432.003.0017BibTeX
@inproceedings{desalvobraz2005ijcai-lifted,
title = {{Lifted First-Order Probabilistic Inference}},
author = {de Salvo Braz, Rodrigo and Amir, Eyal and Roth, Dan},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2005},
pages = {1319-1325},
doi = {10.7551/mitpress/7432.003.0017},
url = {https://mlanthology.org/ijcai/2005/desalvobraz2005ijcai-lifted/}
}