DC Field | Value | Language |
---|---|---|
dc.contributor.author | Todorović, Branimir | en |
dc.contributor.author | Stanković, Miomir | en |
dc.contributor.author | Moraga, Claudio | en |
dc.date.accessioned | 2020-12-11T13:04:31Z | - |
dc.date.available | 2020-12-11T13:04:31Z | - |
dc.date.issued | 2014-01-01 | en |
dc.identifier.isbn | 978-989-758-054-3 | en |
dc.identifier.uri | http://researchrepository.mi.sanu.ac.rs/handle/123456789/4335 | - |
dc.description.abstract | The problem of recurrent neural network training is considered here as an approximate joint Bayesian estimation of the neuron outputs and unknown synaptic weights. We have implemented recursive estimators using nonlinear derivative free approximation of neural network dynamics. The computational efficiency and performances of proposed algorithms as training algorithms for different recurrent neural network architectures are compared on the problem of long term, chaotic time series prediction. | en |
dc.relation.ispartof | NCTA 2014 - Proceedings of the International Conference on Neural Computation Theory and Applications | en |
dc.subject | Bayesian estimation | Chaotic time series prediction | Nonlinear derivative free estimation | Recurrent neural networks | en |
dc.title | Derivative free training of recurrent neural networks a comparison of algorithms and architectures | en |
dc.type | Conference Paper | en |
dc.identifier.doi | 10.5220/0005081900760084 | en |
dc.identifier.scopus | 2-s2.0-84908885128 | en |
dc.relation.firstpage | 76 | en |
dc.relation.lastpage | 84 | en |
dc.description.rank | M33 | - |
item.grantfulltext | none | - |
item.cerifentitytype | Publications | - |
item.fulltext | No Fulltext | - |
item.openairecristype | http://purl.org/coar/resource_type/c_18cf | - |
item.openairetype | Conference Paper | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.