Knowledge representation and reasoning: Difference between revisions

Revision as of 01:50, 14 April 2025

Revision as of 02:17, 14 April 2025

Line 6:

From a broader perspective, parameterized mechanisms of knowledge representation — including [[Neural network (machine learning)|neural network]] architectures such as [[convolutional neural network]]<nowiki/>s and [[Transformer (deep learning architecture)|transformer]]<nowiki/>s — can also be viewed as a type of knowledge representation formalisms. In AI systems, the question of which knowledge representation formalism is most appropriate has long been a subject of extensive debate. For instance, van Harmelen et al. discussed the suitability of logic as a knowledge representation formalism and reviewed arguments presented by anti-logicists<ref>{{Cite book |last=Porter |first=Bruce |title=Handbook of knowledge representation |last2=Lifschitz |first2=Vladimir |last3=Van Harmelen |first3=Frank |date=2008 |publisher=Elsevier |isbn=978-0-444-52211-5 |edition=1st ed |series=Foundations of artificial intelligence |___location=Amsterdam Boston}}</ref>. Smolensky criticized the limitations of symbolic knowledge representation and explored the possibilities of integrating it with connectionist approaches<ref>{{Cite journal |last=Smolensky |first=Paul |date=1988-03 |title=On the proper treatment of connectionism |url=https://www.cambridge.org/core/product/identifier/S0140525X00052432/type/journal_article |journal=Behavioral and Brain Sciences |language=en |volume=11 |issue=1 |pages=1–23 |doi=10.1017/S0140525X00052432 |issn=0140-525X}}</ref>.

InterestinglyRecently, Zhang et al. have recently demonstrated that all universal (or equally expressive and natural) formalisms of knowledge representation are recursively isomorphic<ref>{{Cite journal |last=Zhang |first=Heng |last2=Jiang |first2=Guifei |last3=Quan |first3=Donghui |date=2025-04-11 |title=A Theory of Formalisms for Representing Knowledge |url=https://ojs.aaai.org/index.php/AAAI/article/view/33674 |journal=Proceedings of the AAAI Conference on Artificial Intelligence |language=en |volume=39 |issue=14 |pages=15257–15264 |doi=10.1609/aaai.v39i14.33674 |issn=2374-3468}}</ref>. This result indicates that mainstream knowledge representation formalisms are essentially equivalent in terms of their capacity for realizing [[artificial general intelligence]] (AGI); diverse technical approaches can draw insights from one another via recursive isomorphisms, yet the fundamental challenges are inherently shared.

== History ==