Publikation: Spin Glass Systems as Collective Active Inference
Dateien
Datum
Autor:innen
Herausgeber:innen
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
DOI (zitierfähiger Link)
Internationale Patentnummer
Angaben zur Forschungsförderung
Projekt
Open Access-Veröffentlichung
Core Facility der Universität Konstanz
Titel in einer weiteren Sprache
Publikationstyp
Publikationsstatus
Erschienen in
Zusammenfassung
An open question in the study of emergent behaviour in multi-agent Bayesian systems is the relationship, if any, between individual and collective inference. In this paper we explore the correspondence between generative models that exist at two distinct scales, using spin glass models as a sandbox system to investigate this question. We show that the collective dynamics of a specific type of active inference agent is equivalent to sampling from the stationary distribution of a spin glass system. A collective of specifically-designed active inference agents can thus be described as implementing a form of sampling-based inference (namely, from a Boltzmann machine) at the higher level. However, this equivalence is very fragile, breaking upon simple modifications to the generative models of the individual agents or the nature of their interactions. We discuss the implications of this correspondence and its fragility for the study of multiscale systems composed of Bayesian agents.
Zusammenfassung in einer weiteren Sprache
Fachgebiet (DDC)
Schlagwörter
Konferenz
Rezension
Zitieren
ISO 690
HEINS, Conor, Brennan KLEIN, Daphne DEMEKAS, Miguel AGUILERA, Christopher L. BUCKLEY, 2023. Spin Glass Systems as Collective Active Inference. 3rd International Workshop on Active Inference (IWAI 2022). Grenoble, France, 19. Sept. 2022. In: BUCKLEY, Christopher L., ed. and others. Active Inference : Third International Workshop, IWAI 2022, Grenoble, France, September 19, 2022, Revised Selected Papers. Cham: Springer, 2023, pp. 75-98. Communications in Computer and Information Science. 1721. ISSN 1865-0929. eISSN 1865-0937. ISBN 978-3-031-28718-3. Available under: doi: 10.1007/978-3-031-28719-0_6BibTex
@inproceedings{Heins2023Glass-67352, year={2023}, doi={10.1007/978-3-031-28719-0_6}, title={Spin Glass Systems as Collective Active Inference}, number={1721}, isbn={978-3-031-28718-3}, issn={1865-0929}, publisher={Springer}, address={Cham}, series={Communications in Computer and Information Science}, booktitle={Active Inference : Third International Workshop, IWAI 2022, Grenoble, France, September 19, 2022, Revised Selected Papers}, pages={75--98}, editor={Buckley, Christopher L.}, author={Heins, Conor and Klein, Brennan and Demekas, Daphne and Aguilera, Miguel and Buckley, Christopher L.} }
RDF
<rdf:RDF xmlns:dcterms="http://purl.org/dc/terms/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:bibo="http://purl.org/ontology/bibo/" xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#" xmlns:foaf="http://xmlns.com/foaf/0.1/" xmlns:void="http://rdfs.org/ns/void#" xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/67352"> <dc:contributor>Demekas, Daphne</dc:contributor> <dc:creator>Heins, Conor</dc:creator> <dcterms:issued>2023</dcterms:issued> <dc:creator>Demekas, Daphne</dc:creator> <dc:creator>Buckley, Christopher L.</dc:creator> <dc:contributor>Heins, Conor</dc:contributor> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2023-07-11T07:03:04Z</dc:date> <foaf:homepage rdf:resource="http://localhost:8080/"/> <dc:contributor>Aguilera, Miguel</dc:contributor> <dc:creator>Klein, Brennan</dc:creator> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2023-07-11T07:03:04Z</dcterms:available> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/43615"/> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/43615"/> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/28"/> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/28"/> <dc:language>eng</dc:language> <dc:creator>Aguilera, Miguel</dc:creator> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/67352"/> <dc:contributor>Buckley, Christopher L.</dc:contributor> <dcterms:title>Spin Glass Systems as Collective Active Inference</dcterms:title> <dc:contributor>Klein, Brennan</dc:contributor> <dcterms:abstract>An open question in the study of emergent behaviour in multi-agent Bayesian systems is the relationship, if any, between individual and collective inference. In this paper we explore the correspondence between generative models that exist at two distinct scales, using spin glass models as a sandbox system to investigate this question. We show that the collective dynamics of a specific type of active inference agent is equivalent to sampling from the stationary distribution of a spin glass system. A collective of specifically-designed active inference agents can thus be described as implementing a form of sampling-based inference (namely, from a Boltzmann machine) at the higher level. However, this equivalence is very fragile, breaking upon simple modifications to the generative models of the individual agents or the nature of their interactions. We discuss the implications of this correspondence and its fragility for the study of multiscale systems composed of Bayesian agents.</dcterms:abstract> </rdf:Description> </rdf:RDF>