File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A novel neural network for associative memory via dynamical systems

TitleA novel neural network for associative memory via dynamical systems
Authors
KeywordsAssociative memory
Asymptotic stability
Learning and forgetting algorithm
Neural network
Prototype pattern
Spurious state
Issue Date2006
PublisherAmerican Institute of Mathematical Sciences. The Journal's web site is located at http://www.aimsciences.org/dcdsB.htm
Citation
Discrete And Continuous Dynamical Systems - Series B, 2006, v. 6 n. 3, p. 573-590 How to Cite?
AbstractThis paper proposes a novel neural network model for associative memory using dynamical systems. The proposed model is based on synthesizing the external input vector, which is different from the conventional approach where the design is based on synthesizing the connection matrix. It is shown that this new neural network (a) stores the desired prototype patterns as asymptotically stable equilibrium points, (b) has no spurious states, and (c) has learning and forgetting capabilities. Moreover, new learning and forgetting algorithms are also developed via a novel operation on the matrix space. Numerical examples are presented to illustrate the effectiveness of the proposed neural network for associative memory. Indeed, results of simulation experiments demonstrate that the neural network is effective and can be implemented easily.
Persistent Identifierhttp://hdl.handle.net/10722/74516
ISSN
2023 Impact Factor: 1.3
2023 SCImago Journal Rankings: 0.655
References

 

DC FieldValueLanguage
dc.contributor.authorMak, KLen_HK
dc.contributor.authorPeng, JGen_HK
dc.contributor.authorXu, ZBen_HK
dc.contributor.authorYiu, KFCen_HK
dc.date.accessioned2010-09-06T07:02:06Z-
dc.date.available2010-09-06T07:02:06Z-
dc.date.issued2006en_HK
dc.identifier.citationDiscrete And Continuous Dynamical Systems - Series B, 2006, v. 6 n. 3, p. 573-590en_HK
dc.identifier.issn1531-3492en_HK
dc.identifier.urihttp://hdl.handle.net/10722/74516-
dc.description.abstractThis paper proposes a novel neural network model for associative memory using dynamical systems. The proposed model is based on synthesizing the external input vector, which is different from the conventional approach where the design is based on synthesizing the connection matrix. It is shown that this new neural network (a) stores the desired prototype patterns as asymptotically stable equilibrium points, (b) has no spurious states, and (c) has learning and forgetting capabilities. Moreover, new learning and forgetting algorithms are also developed via a novel operation on the matrix space. Numerical examples are presented to illustrate the effectiveness of the proposed neural network for associative memory. Indeed, results of simulation experiments demonstrate that the neural network is effective and can be implemented easily.en_HK
dc.languageengen_HK
dc.publisherAmerican Institute of Mathematical Sciences. The Journal's web site is located at http://www.aimsciences.org/dcdsB.htmen_HK
dc.relation.ispartofDiscrete and Continuous Dynamical Systems - Series Ben_HK
dc.subjectAssociative memoryen_HK
dc.subjectAsymptotic stabilityen_HK
dc.subjectLearning and forgetting algorithmen_HK
dc.subjectNeural networken_HK
dc.subjectPrototype patternen_HK
dc.subjectSpurious stateen_HK
dc.titleA novel neural network for associative memory via dynamical systemsen_HK
dc.typeArticleen_HK
dc.identifier.openurlhttp://library.hku.hk:4550/resserv?sid=HKU:IR&issn=1531-3492&volume=6&issue=3&spage=573&epage=590&date=2006&atitle=A+novel+neural+network+for+associative+memory+via+dynamical+systemsen_HK
dc.identifier.emailMak, KL:makkl@hkucc.hku.hken_HK
dc.identifier.emailYiu, KFC:cedric@hkucc.hku.hken_HK
dc.identifier.authorityMak, KL=rp00154en_HK
dc.identifier.authorityYiu, KFC=rp00206en_HK
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.3934/dcdsb.2006.6.573-
dc.identifier.scopuseid_2-s2.0-33745257609en_HK
dc.identifier.hkuros124167en_HK
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-33745257609&selection=ref&src=s&origin=recordpageen_HK
dc.identifier.volume6en_HK
dc.identifier.issue3en_HK
dc.identifier.spage573en_HK
dc.identifier.epage590en_HK
dc.publisher.placeUnited Statesen_HK
dc.identifier.scopusauthoridMak, KL=7102680226en_HK
dc.identifier.scopusauthoridPeng, JG=7401959175en_HK
dc.identifier.scopusauthoridXu, ZB=7405426248en_HK
dc.identifier.scopusauthoridYiu, KFC=24802813000en_HK
dc.identifier.issnl1531-3492-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats