Some advances regarding ontologies and neuro-symbolic artificial intelligence

Pascal Hitzler
ECMLPKDD Workshop on Meta-Knowledge Transfer, PMLR 191:8-10, 2022.

Abstract

This abstract serves as pointers to the most relevant literature references underlying my workshop keynote. Symbolic AI (based on knowledge representation and formal logic) and AI based on artificial neural networks (such as deep learning) are fundamentally different approaches to artificial intelligence with complementary capabilities. The former are transparent and data-efficient, but they are sensitive to noise and cannot be applied to non-symbolic domains where the data is ambiguous. The latter can learn complex tasks from examples, are robust to noise, but are black boxes; require large amounts of – not necessarily easily obtained – data, and are slow to learn and prone to adversarial examples. Either paradigm excels at certain types of problems where the other paradigm performs poorly. In order to develop stronger AI systems, integrated neuro-symbolic systems that combine artificial neural networks and symbolic reasoning are being sought. In this talk, we discuss two related lines of investigation in neuro-symbolic AI. (1) We report on our work in progress of using concept induction over ontologies for explaining deep learning systems. (2) We present recent results regarding the acquisition of formal logical reasoning capabilities over ontologies, though deep learning, which we call Deep Deductive Reasoning.

Cite this Paper

BibTeX
@InProceedings{pmlr-v191-hitzler22a, title = {Some advances regarding ontologies and neuro-symbolic artificial intelligence}, author = {Hitzler, Pascal}, booktitle = {ECMLPKDD Workshop on Meta-Knowledge Transfer}, pages = {8--10}, year = {2022}, editor = {Brazdil, Pavel and van Rijn, Jan N. and Gouk, Henry and Mohr, Felix}, volume = {191}, series = {Proceedings of Machine Learning Research}, month = {23 Sep}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v191/hitzler22a/hitzler22a.pdf}, url = {https://proceedings.mlr.press/v191/hitzler22a.html}, abstract = {This abstract serves as pointers to the most relevant literature references underlying my workshop keynote. Symbolic AI (based on knowledge representation and formal logic) and AI based on artificial neural networks (such as deep learning) are fundamentally different approaches to artificial intelligence with complementary capabilities. The former are transparent and data-efficient, but they are sensitive to noise and cannot be applied to non-symbolic domains where the data is ambiguous. The latter can learn complex tasks from examples, are robust to noise, but are black boxes; require large amounts of – not necessarily easily obtained – data, and are slow to learn and prone to adversarial examples. Either paradigm excels at certain types of problems where the other paradigm performs poorly. In order to develop stronger AI systems, integrated neuro-symbolic systems that combine artificial neural networks and symbolic reasoning are being sought. In this talk, we discuss two related lines of investigation in neuro-symbolic AI. (1) We report on our work in progress of using concept induction over ontologies for explaining deep learning systems. (2) We present recent results regarding the acquisition of formal logical reasoning capabilities over ontologies, though deep learning, which we call Deep Deductive Reasoning. } }
Endnote
%0 Conference Paper %T Some advances regarding ontologies and neuro-symbolic artificial intelligence %A Pascal Hitzler %B ECMLPKDD Workshop on Meta-Knowledge Transfer %C Proceedings of Machine Learning Research %D 2022 %E Pavel Brazdil %E Jan N. van Rijn %E Henry Gouk %E Felix Mohr %F pmlr-v191-hitzler22a %I PMLR %P 8--10 %U https://proceedings.mlr.press/v191/hitzler22a.html %V 191 %X This abstract serves as pointers to the most relevant literature references underlying my workshop keynote. Symbolic AI (based on knowledge representation and formal logic) and AI based on artificial neural networks (such as deep learning) are fundamentally different approaches to artificial intelligence with complementary capabilities. The former are transparent and data-efficient, but they are sensitive to noise and cannot be applied to non-symbolic domains where the data is ambiguous. The latter can learn complex tasks from examples, are robust to noise, but are black boxes; require large amounts of – not necessarily easily obtained – data, and are slow to learn and prone to adversarial examples. Either paradigm excels at certain types of problems where the other paradigm performs poorly. In order to develop stronger AI systems, integrated neuro-symbolic systems that combine artificial neural networks and symbolic reasoning are being sought. In this talk, we discuss two related lines of investigation in neuro-symbolic AI. (1) We report on our work in progress of using concept induction over ontologies for explaining deep learning systems. (2) We present recent results regarding the acquisition of formal logical reasoning capabilities over ontologies, though deep learning, which we call Deep Deductive Reasoning.
APA
Hitzler, P.. (2022). Some advances regarding ontologies and neuro-symbolic artificial intelligence. ECMLPKDD Workshop on Meta-Knowledge Transfer, in Proceedings of Machine Learning Research 191:8-10 Available from https://proceedings.mlr.press/v191/hitzler22a.html.

Related Material