PAC-Reasoning in Relational Domains
Artificial Intelligence
2018-07-05 v3 Machine Learning
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
@article{arxiv.1803.05768,
title = {PAC-Reasoning in Relational Domains},
author = {Ondrej Kuzelka and Yuyi Wang and Jesse Davis and Steven Schockaert},
journal= {arXiv preprint arXiv:1803.05768},
year = {2018}
}
Comments
Longer version of paper appearing in UAI 2018