Show simple item record

2020-06-07Zeitschriftenartikel DOI: 10.18452/25019
Intrinsic motivation and episodic memories for robot exploration of high-dimensional sensory spaces
dc.contributor.authorSchillaci, Guido
dc.contributor.authorPico Villalpando, Antonio
dc.contributor.authorHafner, Verena
dc.contributor.authorHanappe, Peter
dc.contributor.authorColliaux, David
dc.contributor.authorWintz, Timothée
dc.date.accessioned2022-07-18T11:49:22Z
dc.date.available2022-07-18T11:49:22Z
dc.date.issued2020-06-07none
dc.date.updated2021-12-07T23:34:01Z
dc.identifier.urihttp://edoc.hu-berlin.de/18452/25697
dc.descriptionDieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer überregionalen Konsortiallizenz frei zugänglich.none
dc.description.abstractThis work presents an architecture that generates curiosity-driven goal-directed exploration behaviours for an image sensor of a microfarming robot. A combination of deep neural networks for offline unsupervised learning of low-dimensional features from images and of online learning of shallow neural networks representing the inverse and forward kinematics of the system have been used. The artificial curiosity system assigns interest values to a set of pre-defined goals and drives the exploration towards those that are expected to maximise the learning progress. We propose the integration of an episodic memory in intrinsic motivation systems to face catastrophic forgetting issues, typically experienced when performing online updates of artificial neural networks. Our results show that adopting an episodic memory system not only prevents the computational models from quickly forgetting knowledge that has been previously acquired but also provides new avenues for modulating the balance between plasticity and stability of the models.eng
dc.description.sponsorshipH2020 Marie Skłodowska-Curie Actionshttps://doi.org/10.13039/100010665
dc.description.sponsorshipHorizon 2020 Framework Programmehttps://doi.org/10.13039/100010661
dc.language.isoengnone
dc.publisherHumboldt-Universität zu Berlin
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/
dc.subjectAdaptive modelseng
dc.subjectpredictive modelseng
dc.subjectepisodic memoryeng
dc.subjectmemory consolidationeng
dc.subjectintrinsic motivationeng
dc.subjectroboticseng
dc.subject.ddc004 Informatiknone
dc.titleIntrinsic motivation and episodic memories for robot exploration of high-dimensional sensory spacesnone
dc.typearticle
dc.identifier.urnurn:nbn:de:kobv:11-110-18452/25697-8
dc.identifier.doihttp://dx.doi.org/10.18452/25019
dc.type.versionpublishedVersionnone
local.edoc.pages18none
local.edoc.type-nameZeitschriftenartikel
local.edoc.container-typeperiodical
local.edoc.container-type-nameZeitschrift
local.edoc.container-year2021none
dc.description.versionPeer Reviewednone
dc.identifier.eissn1741-2633
dcterms.bibliographicCitation.doi10.1177/1059712320922916none
dcterms.bibliographicCitation.journaltitleAdaptive behaviornone
dcterms.bibliographicCitation.volume29none
dcterms.bibliographicCitation.issue6none
dcterms.bibliographicCitation.originalpublishernameSagenone
dcterms.bibliographicCitation.originalpublisherplaceThousand Oaks, Calif.none
dcterms.bibliographicCitation.pagestart549none
dcterms.bibliographicCitation.pageend566none
bua.departmentMathematisch-Naturwissenschaftliche Fakultätnone

Show simple item record