Neural markov logic networks

Giuseppe Marra, Ondřej Kuželka
Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, PMLR 161:908-917, 2021.

Abstract

We introduce neural Markov logic networks (NMLNs), a statistical relational learning system that borrows ideas from Markov logic. Like Markov logic networks (MLNs), NMLNs are an exponential-family model for modelling distributions over possible worlds, but unlike MLNs, they do not rely on explicitly specified first-order logic rules. Instead, NMLNs learn an implicit representation of such rules as a neural network that acts as a potential function on fragments of the relational structure. Similarly to many neural symbolic methods, NMLNs can exploit embeddings of constants but, unlike them, NMLNs work well also in their absence. This is extremely important for predicting in settings other than the transductive one. We showcase the potential of NMLNs on knowledge-base completion, triple classification and on generation of molecular (graph) data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v161-marra21a, title = {Neural markov logic networks}, author = {Marra, Giuseppe and Ku\v{z}elka, Ond\v{r}ej}, booktitle = {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence}, pages = {908--917}, year = {2021}, editor = {de Campos, Cassio and Maathuis, Marloes H.}, volume = {161}, series = {Proceedings of Machine Learning Research}, month = {27--30 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v161/marra21a/marra21a.pdf}, url = {https://proceedings.mlr.press/v161/marra21a.html}, abstract = {We introduce neural Markov logic networks (NMLNs), a statistical relational learning system that borrows ideas from Markov logic. Like Markov logic networks (MLNs), NMLNs are an exponential-family model for modelling distributions over possible worlds, but unlike MLNs, they do not rely on explicitly specified first-order logic rules. Instead, NMLNs learn an implicit representation of such rules as a neural network that acts as a potential function on fragments of the relational structure. Similarly to many neural symbolic methods, NMLNs can exploit embeddings of constants but, unlike them, NMLNs work well also in their absence. This is extremely important for predicting in settings other than the transductive one. We showcase the potential of NMLNs on knowledge-base completion, triple classification and on generation of molecular (graph) data.} }
Endnote
%0 Conference Paper %T Neural markov logic networks %A Giuseppe Marra %A Ondřej Kuželka %B Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2021 %E Cassio de Campos %E Marloes H. Maathuis %F pmlr-v161-marra21a %I PMLR %P 908--917 %U https://proceedings.mlr.press/v161/marra21a.html %V 161 %X We introduce neural Markov logic networks (NMLNs), a statistical relational learning system that borrows ideas from Markov logic. Like Markov logic networks (MLNs), NMLNs are an exponential-family model for modelling distributions over possible worlds, but unlike MLNs, they do not rely on explicitly specified first-order logic rules. Instead, NMLNs learn an implicit representation of such rules as a neural network that acts as a potential function on fragments of the relational structure. Similarly to many neural symbolic methods, NMLNs can exploit embeddings of constants but, unlike them, NMLNs work well also in their absence. This is extremely important for predicting in settings other than the transductive one. We showcase the potential of NMLNs on knowledge-base completion, triple classification and on generation of molecular (graph) data.
APA
Marra, G. & Kuželka, O.. (2021). Neural markov logic networks. Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 161:908-917 Available from https://proceedings.mlr.press/v161/marra21a.html.

Related Material