PAC-Reasoning in Relational Domains
Abstract
We consider the problem of predicting plausible missing facts in relational data, given a set of imperfect logical rules. In particular, our aim is to provide bounds on the (expected) number of incorrect inferences that are made in this way. Since for classical inference it is in general impossible to bound this number in a non-trivial way, we consider two inference relations that weaken, but remain close in spirit to classical inference.
Cite
Text
Kuzelka et al. "PAC-Reasoning in Relational Domains." Conference on Uncertainty in Artificial Intelligence, 2018.Markdown
[Kuzelka et al. "PAC-Reasoning in Relational Domains." Conference on Uncertainty in Artificial Intelligence, 2018.](https://mlanthology.org/uai/2018/kuzelka2018uai-pac/)BibTeX
@inproceedings{kuzelka2018uai-pac,
title = {{PAC-Reasoning in Relational Domains}},
author = {Kuzelka, Ondrej and Wang, Yuyi and Davis, Jesse and Schockaert, Steven},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2018},
pages = {927-936},
url = {https://mlanthology.org/uai/2018/kuzelka2018uai-pac/}
}