Fast Learning of Relational Kernels
Abstract
We develop a general theoretical framework for statistical logical learning with kernels based on dynamic propositionalization, where structure learning corresponds to inferring a suitable kernel on logical objects, and parameter learning corresponds to function learning in the resulting reproducing kernel Hilbert space. In particular, we study the case where structure learning is performed by a simple FOIL-like algorithm, and propose alternative scoring functions for guiding the search process. We present an empirical evaluation on several data sets in the single-task as well as in the multi-task setting.
Cite
Text
Landwehr et al. "Fast Learning of Relational Kernels." Machine Learning, 2010. doi:10.1007/S10994-009-5163-1Markdown
[Landwehr et al. "Fast Learning of Relational Kernels." Machine Learning, 2010.](https://mlanthology.org/mlj/2010/landwehr2010mlj-fast/) doi:10.1007/S10994-009-5163-1BibTeX
@article{landwehr2010mlj-fast,
title = {{Fast Learning of Relational Kernels}},
author = {Landwehr, Niels and Passerini, Andrea and De Raedt, Luc and Frasconi, Paolo},
journal = {Machine Learning},
year = {2010},
pages = {305-342},
doi = {10.1007/S10994-009-5163-1},
volume = {78},
url = {https://mlanthology.org/mlj/2010/landwehr2010mlj-fast/}
}