Simulation of Entanglement between Two AIs: A Proof of Concept for Ontology Matching

This paper proposes a computational simulation of the concept of entanglement between two artificial agents, applied as a proof of concept for ontology matching. The simulation, to be developed in a future implementation stage, will instantiate two independent AI agents, each representing one of the two ontologies to be matched, and each operating in its own private reference frame generated by orthogonal random matrices with distinct seeds. Because the two reference frames are incommensurable, the agents are, by construction, uncorrelated: the expected cosine similarity between any pair of representations is zero. A linear channel (ridge regression codec) will then be trained on a small set of known correspondence pairs, analogous to calibration pairs, and is expected to align the two reference frames on held-out pairs, while a randomized control set should remain at noise level. We formalize this setup through a diagnosis protocol that will report (i) the alignment residual against the theoretical optimal frame matcher, (ii) the cosine similarity before and after the channel, and (iii) the gap between correlated pairs and a random control. The proof of concept will be carried out on the Anatomy track of the OAEI, using its public gold standard, in order to demonstrate methodological feasibility. The intended result is a reproducible, locally executable demonstration that learning a mapping between two agents' representations is mathematically equivalent to concentrating correlation on a low-dimensional subspace, the computational analogue of partial entanglement, and that this learned channel can support automated ontology matching. The program is still under development and will be made available upon completion.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-13
DOI
https://doi.org/10.5281/zenodo.24180151
Primary Topic
Semantic Web and Ontologies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Simulation of Entanglement between Two AIs: A Proof of Concept for Ontology Matching

Roberto Monteiro Dias
Zenodo (CERN European Organization for Nuclear Research)
Semantic Web and Ontologies
preprint

Simulation of Entanglement between Two AIs: A Proof of Concept for Ontology Matching

Roberto Monteiro Dias
preprint en

Abstract

This paper proposes a computational simulation of the concept of entanglement between two artificial agents, applied as a proof of concept for ontology matching. The simulation, to be developed in a future implementation stage, will instantiate two independent AI agents, each representing one of the two ontologies to be matched, and each operating in its own private reference frame generated by orthogonal random matrices with distinct seeds. Because the two reference frames are incommensurable, the agents are, by construction, uncorrelated: the expected cosine similarity between any pair of representations is zero. A linear channel (ridge regression codec) will then be trained on a small set of known correspondence pairs, analogous to calibration pairs, and is expected to align the two reference frames on held-out pairs, while a randomized control set should remain at noise level. We formalize this setup through a diagnosis protocol that will report (i) the alignment residual against the theoretical optimal frame matcher, (ii) the cosine similarity before and after the channel, and (iii) the gap between correlated pairs and a random control. The proof of concept will be carried out on the Anatomy track of the OAEI, using its public gold standard, in order to demonstrate methodological feasibility. The intended result is a reproducible, locally executable demonstration that learning a mapping between two agents' representations is mathematically equivalent to concentrating correlation on a low-dimensional subspace, the computational analogue of partial entanglement, and that this learned channel can support automated ontology matching. The program is still under development and will be made available upon completion.

Zenodo (CERN European Organization for Nuclear Research)
Universidade Federal do Estado do Rio de Janeiro (BR)
Semantic Web and Ontologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon