Projective simulation with generalization

Cite This

Files in this item

Checksum: MD5:46cd9611621c01dfab412d842fbe175e

MELNIKOV, Alexey A, Adi MAKMAL, Vedran DUNJKO, Hans BRIEGEL, 2017. Projective simulation with generalization. In: Scientific reports. 7(1), 14430. eISSN 2045-2322. Available under: doi: 10.1038/s41598-017-14740-y

@article{Melnikov2017Proje-40977, title={Projective simulation with generalization}, year={2017}, doi={10.1038/s41598-017-14740-y}, number={1}, volume={7}, journal={Scientific reports}, author={Melnikov, Alexey A and Makmal, Adi and Dunjko, Vedran and Briegel, Hans}, note={Article Number: 14430} }

<rdf:RDF xmlns:dcterms="" xmlns:dc="" xmlns:rdf="" xmlns:bibo="" xmlns:dspace="" xmlns:foaf="" xmlns:void="" xmlns:xsd="" > <rdf:Description rdf:about=""> <dspace:isPartOfCollection rdf:resource=""/> <dcterms:abstract xml:lang="eng">The ability to generalize is an important feature of any intelligent agent. Not only because it may allow the agent to cope with large amounts of data, but also because in some environments, an agent with no generalization capabilities cannot learn. In this work we outline several criteria for generalization, and present a dynamic and autonomous machinery that enables projective simulation agents to meaningfully generalize. Projective simulation, a novel, physical approach to artificial intelligence, was recently shown to perform well in standard reinforcement learning problems, with applications in advanced robotics as well as quantum experiments. Both the basic projective simulation model and the presented generalization machinery are based on very simple principles. This allows us to provide a full analytical analysis of the agent's performance and to illustrate the benefit the agent gains by generalizing. Specifically, we show that already in basic (but extreme) environments, learning without generalization may be impossible, and demonstrate how the presented generalization machinery enables the projective simulation agent to learn.</dcterms:abstract> <dc:rights>terms-of-use</dc:rights> <dcterms:available rdf:datatype="">2017-12-20T09:15:35Z</dcterms:available> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> <dspace:hasBitstream rdf:resource=""/> <bibo:uri rdf:resource=""/> <dc:contributor>Briegel, Hans</dc:contributor> <dc:creator>Makmal, Adi</dc:creator> <dc:creator>Briegel, Hans</dc:creator> <dcterms:hasPart rdf:resource=""/> <dc:date rdf:datatype="">2017-12-20T09:15:35Z</dc:date> <dc:language>eng</dc:language> <dcterms:rights rdf:resource=""/> <dc:creator>Melnikov, Alexey A</dc:creator> <dc:creator>Dunjko, Vedran</dc:creator> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dcterms:title>Projective simulation with generalization</dcterms:title> <dc:contributor>Melnikov, Alexey A</dc:contributor> <dc:contributor>Makmal, Adi</dc:contributor> <dc:contributor>Dunjko, Vedran</dc:contributor> <dcterms:issued>2017</dcterms:issued> <dcterms:isPartOf rdf:resource=""/> </rdf:Description> </rdf:RDF>

Downloads since Dec 20, 2017 (Information about access statistics)

Melnikov_2-12vype5f9hshs2.pdf 121

This item appears in the following Collection(s)

Search KOPS


My Account