VC-Dimension Based Generalization Bounds for Relational Learning
Abstract
In many applications of relational learning, the available data can be seen as a sample from a larger relational structure (e.g. we may be given a small fragment from some social network). In this paper we are particularly concerned with scenarios in which we can assume that (i) the domain elements appearing in the given sample have been uniformly sampled without replacement from the (unknown) full domain and (ii) the sample is complete for these domain elements (i.e. it is the full substructure induced by these elements). Within this setting, we study bounds on the error of sufficient statistics of relational models that are estimated on the available data. As our main result, we prove a bound based on a variant of the Vapnik-Chervonenkis dimension which is suitable for relational data.
Cite
Text
Kuzelka et al. "VC-Dimension Based Generalization Bounds for Relational Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2018. doi:10.1007/978-3-030-10928-8_16Markdown
[Kuzelka et al. "VC-Dimension Based Generalization Bounds for Relational Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2018.](https://mlanthology.org/ecmlpkdd/2018/kuzelka2018ecmlpkdd-vcdimension/) doi:10.1007/978-3-030-10928-8_16BibTeX
@inproceedings{kuzelka2018ecmlpkdd-vcdimension,
title = {{VC-Dimension Based Generalization Bounds for Relational Learning}},
author = {Kuzelka, Ondrej and Wang, Yuyi and Schockaert, Steven},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2018},
pages = {259-275},
doi = {10.1007/978-3-030-10928-8_16},
url = {https://mlanthology.org/ecmlpkdd/2018/kuzelka2018ecmlpkdd-vcdimension/}
}