Ontologies represent human domain expertise symbolically in a way that is accessible to human experts and suitable for a variety of applications. As a result, they are widely used in scientific research. A challenge for the neuro-symbolic field is how to use the knowledge encoded in ontologies together with sub-symbolic learning approaches. In this chapter we describe a general neuro-symbolic architecture for using knowledge from ontologies to improve the generalisability and accuracy of predictions of a deep neural network applied to chemical data. The architecture consists of a multi-layer network with a multi-step training process: first, the network is given a self-supervised pre-training step with a masked language step in order to learn the input representation. Second, the network is given an ontology pre-training step in which the network learns to predict membership in the classes of the ontology as a way to learn organising knowledge from the ontology. Finally, we show that visualisation of the attention weights of the ontology-trained network allows some form of interpretability of network predictions. In general, we propose a three-layered architecture for neuro-symbolic integration, with layers for 1) encoding, 2) ontological classification, and 3) ontology-driven logical loss.
%0 Book Section
%1 glauer2023neurosymbolic
%A Glauer, Martin
%A Mossakowski, Till
%A Neuhaus, Fabian
%A Memariani, Adel
%A Hastings, Janna
%B A Compendium of Neuro-Symbolic Artificial Intelligence
%D 2023
%E Hitzler, Pascal
%E Sarker, Md Kamruzzaman
%E Eberhart, Aaron
%I IOS press
%K chemistry myown neuro-symbolic ontology
%P 460 - 484
%R 10.3233/FAIA230153
%T Neuro-symbolic semantic learning for chemistry
%U https://ebooks.iospress.nl/volumearticle/63730
%V 369
%X Ontologies represent human domain expertise symbolically in a way that is accessible to human experts and suitable for a variety of applications. As a result, they are widely used in scientific research. A challenge for the neuro-symbolic field is how to use the knowledge encoded in ontologies together with sub-symbolic learning approaches. In this chapter we describe a general neuro-symbolic architecture for using knowledge from ontologies to improve the generalisability and accuracy of predictions of a deep neural network applied to chemical data. The architecture consists of a multi-layer network with a multi-step training process: first, the network is given a self-supervised pre-training step with a masked language step in order to learn the input representation. Second, the network is given an ontology pre-training step in which the network learns to predict membership in the classes of the ontology as a way to learn organising knowledge from the ontology. Finally, we show that visualisation of the attention weights of the ontology-trained network allows some form of interpretability of network predictions. In general, we propose a three-layered architecture for neuro-symbolic integration, with layers for 1) encoding, 2) ontological classification, and 3) ontology-driven logical loss.
%& 21
@incollection{glauer2023neurosymbolic,
abstract = {Ontologies represent human domain expertise symbolically in a way that is accessible to human experts and suitable for a variety of applications. As a result, they are widely used in scientific research. A challenge for the neuro-symbolic field is how to use the knowledge encoded in ontologies together with sub-symbolic learning approaches. In this chapter we describe a general neuro-symbolic architecture for using knowledge from ontologies to improve the generalisability and accuracy of predictions of a deep neural network applied to chemical data. The architecture consists of a multi-layer network with a multi-step training process: first, the network is given a self-supervised pre-training step with a masked language step in order to learn the input representation. Second, the network is given an ontology pre-training step in which the network learns to predict membership in the classes of the ontology as a way to learn organising knowledge from the ontology. Finally, we show that visualisation of the attention weights of the ontology-trained network allows some form of interpretability of network predictions. In general, we propose a three-layered architecture for neuro-symbolic integration, with layers for 1) encoding, 2) ontological classification, and 3) ontology-driven logical loss.},
added-at = {2023-05-27T21:50:55.000+0200},
author = {Glauer, Martin and Mossakowski, Till and Neuhaus, Fabian and Memariani, Adel and Hastings, Janna},
biburl = {https://www.bibsonomy.org/bibtex/2a6e81e065a4589e72684474d485bd3ab/tillmo},
booktitle = {A Compendium of Neuro-Symbolic Artificial Intelligence},
chapter = 21,
doi = {10.3233/FAIA230153},
editor = {Hitzler, Pascal and Sarker, Md Kamruzzaman and Eberhart, Aaron},
interhash = {d9b0ff9f2fcb54585be85987f1186a9b},
intrahash = {a6e81e065a4589e72684474d485bd3ab},
keywords = {chemistry myown neuro-symbolic ontology},
pages = { 460 - 484},
publisher = {IOS press},
series = {Frontiers in Artificial Intelligence and Applications},
timestamp = {2023-08-04T18:19:04.000+0200},
title = {Neuro-symbolic semantic learning for chemistry},
url = {https://ebooks.iospress.nl/volumearticle/63730},
volume = 369,
year = 2023
}