Compromising algorithmicity and plasticity in autonomous agent control architectures: The autonomous cell

Tzafestas, ES

dc.contributor.author	Tzafestas, ES	en
dc.date.accessioned	2014-03-01T01:48:32Z
dc.date.available	2014-03-01T01:48:32Z
dc.date.issued	1999	en
dc.identifier.issn	03341860	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/25509
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-0032658716&partnerID=40&md5=d2e437bb88ffcbbea3f5805f5bd92028	en
dc.subject	Adaptation	en
dc.subject	Algorithmicity	en
dc.subject	Cellularity	en
dc.subject	Fault-tolerance	en
dc.subject	Plasticity	en
dc.subject	Self-organization	en
dc.subject.other	Adaptive systems	en
dc.subject.other	Artificial intelligence	en
dc.subject.other	Buffer storage	en
dc.subject.other	Computer networks	en
dc.subject.other	Computer simulation	en
dc.subject.other	Fault tolerant computer systems	en
dc.subject.other	Navigation systems	en
dc.subject.other	Redundancy	en
dc.subject.other	Adaptation	en
dc.subject.other	Algorithmicity	en
dc.subject.other	Autonomous cell	en
dc.subject.other	Cellularity	en
dc.subject.other	Self organization	en
dc.subject.other	Intelligent control	en
dc.title	Compromising algorithmicity and plasticity in autonomous agent control architectures: The autonomous cell	en
heal.type	journalArticle	en
heal.publicationDate	1999	en
heal.abstract	Whereas algorithmic autonomous agent control architectures demonstrate high efficiency, they suffer from network structure rigidity that shows in the liability to crucial errors. On the other hand, the redundancy inherent in most connectionist architectures allows for continuous self-organization that compensates for limited-scale neuron failures. In this work, we are seeking to compromise algorithmicity and plasticity in front of local network failures, by extending a basic algorithmic cell model. The extension is twofold: on the one hand we introduce motivation to the cell level, which shows as preference to consume some kinds of messages, while on the other hand we introduce sociality, which shows as adaptivity of the cell to the motivations of its neighbors. Unlike usual connectionist models, there are no connections between cells, but message buffers shared by all cells of a level; this way, cells can be viewed as floating in a common interaction medium and competing with one another. Another important feature of this arrangement is the necessity of an immune system, i.e., a population of cells that recognize and eliminate the messages that might be detrimental to the integrity of the cellular agent. The model's self-organizational potential is illustrated on the example case of a navigation system. Our simulation results show that the cellular network exhibit plasticity and recover from various types of failures by ""discovering"" alternative message flow pathways and that multiple failures slow down the system's responsiveness to external events. Issues such as selectivity and the role of diversity are also discussed.	en
heal.journalName	Journal of Intelligent Systems	en
dc.identifier.volume	9	en
dc.identifier.issue	2	en
dc.identifier.spage	135	en
dc.identifier.epage	174	en