FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2554138451> ?p ?o ?g. }

• W2554138451 abstract "In this paper, a variant of Backpropagation algorithm is proposed for feed-forward neural networks learning. The proposed algorithm improve the backpropagation training in terms of quick convergence of the solution depending on the slope of the error graph and increase the speed of convergence of the system. Simulations are conducted to compare and evaluate the convergence behavior and the speed factor of the proposed algorithm with existing Backpropagation algorithm. Simulation results of large-scale classical neural- network benchmarks are presented which reveal the power of the proposed algorithm to obtain actual solutions. derivative based algorithm that have been proposed for the training of feed-forward networks which combines the excellent local convergence properties of Gauss-Newton method near a minimum with the consistent error decrease provided by (a suitably scaled) gradient descent faraway from the solution. In first-order methods (such as gradient descent), a local minimizer problem is overshooted with the inclusion of momentum term. The momentum term actually inserts second-order information in the training process and provides iterations whose form is similar to the conjugate gradient (CG) method. The major difference of Backpropagation with the conjugate gradient method is that the coefficients regulating the weighting between the gradient and the momentum term are heuristically selected in BP, whereas in the CG algorithm these coefficients are adaptively determined. However, these algorithms also share problems (17) present in the standard Backpropagation algorithm and may converge faster in some cases and slower in others. Comparison of the speeds of convergence of different schemes for implementing Backpropagation is not clear-cut, though a discussion on benchmarking of the algorithms can be found (18). In this paper, a proposal for a variant of back-propagation algorithm for FNN with time-varying inputs has been presented which is capable of overcoming the shortcomings of the BP as discussed above. In the experimental section, the proposed algorithm is compared with the existing Backpropagation algorithm for training multilayer feed-forward networks on training tasks that are well known for their complexity. It was observed that the" @default.
• W2554138451 created "2016-11-30" @default.
• W2554138451 creator A5071321435 @default.
• W2554138451 creator A5071610141 @default.
• W2554138451 date "2007-01-01" @default.
• W2554138451 modified "2023-09-27" @default.
• W2554138451 title "A VARIANT OF BACK-PROPAGATION ALGORITHM FOR MULTILAYER FEED-FORWARD NETWORK" @default.
• W2554138451 cites W1482700500 @default.
• W2554138451 cites W1507336107 @default.
• W2554138451 cites W1531857721 @default.
• W2554138451 cites W1532029644 @default.
• W2554138451 cites W1977923618 @default.
• W2554138451 cites W2075194837 @default.
• W2554138451 cites W2077658674 @default.
• W2554138451 cites W2078376347 @default.
• W2554138451 cites W2084606781 @default.
• W2554138451 cites W2096835548 @default.
• W2554138451 cites W2104893957 @default.
• W2554138451 cites W2109239789 @default.
• W2554138451 cites W2109530046 @default.
• W2554138451 cites W2134331967 @default.
• W2554138451 cites W2152868637 @default.
• W2554138451 cites W2155482699 @default.
• W2554138451 cites W2160208155 @default.
• W2554138451 cites W2766736793 @default.
• W2554138451 cites W2796575135 @default.
• W2554138451 cites W2904250082 @default.
• W2554138451 cites W656153502 @default.
• W2554138451 cites W138885771 @default.
• W2554138451 hasPublicationYear "2007" @default.
• W2554138451 type Work @default.
• W2554138451 sameAs 2554138451 @default.
• W2554138451 citedByCount "0" @default.
• W2554138451 crossrefType "journal-article" @default.
• W2554138451 hasAuthorship W2554138451A5071321435 @default.
• W2554138451 hasAuthorship W2554138451A5071610141 @default.
• W2554138451 hasConcept C11413529 @default.
• W2554138451 hasConcept C126838900 @default.
• W2554138451 hasConcept C147168706 @default.
• W2554138451 hasConcept C153258448 @default.
• W2554138451 hasConcept C154945302 @default.
• W2554138451 hasConcept C155032097 @default.
• W2554138451 hasConcept C162324750 @default.
• W2554138451 hasConcept C177973122 @default.
• W2554138451 hasConcept C183115368 @default.
• W2554138451 hasConcept C2777303404 @default.
• W2554138451 hasConcept C33923547 @default.
• W2554138451 hasConcept C41008148 @default.
• W2554138451 hasConcept C50522688 @default.
• W2554138451 hasConcept C50644808 @default.
• W2554138451 hasConcept C71924100 @default.
• W2554138451 hasConcept C81184566 @default.
• W2554138451 hasConcept C98359873 @default.
• W2554138451 hasConceptScore W2554138451C11413529 @default.
• W2554138451 hasConceptScore W2554138451C126838900 @default.
• W2554138451 hasConceptScore W2554138451C147168706 @default.
• W2554138451 hasConceptScore W2554138451C153258448 @default.
• W2554138451 hasConceptScore W2554138451C154945302 @default.
• W2554138451 hasConceptScore W2554138451C155032097 @default.
• W2554138451 hasConceptScore W2554138451C162324750 @default.
• W2554138451 hasConceptScore W2554138451C177973122 @default.
• W2554138451 hasConceptScore W2554138451C183115368 @default.
• W2554138451 hasConceptScore W2554138451C2777303404 @default.
• W2554138451 hasConceptScore W2554138451C33923547 @default.
• W2554138451 hasConceptScore W2554138451C41008148 @default.
• W2554138451 hasConceptScore W2554138451C50522688 @default.
• W2554138451 hasConceptScore W2554138451C50644808 @default.
• W2554138451 hasConceptScore W2554138451C71924100 @default.
• W2554138451 hasConceptScore W2554138451C81184566 @default.
• W2554138451 hasConceptScore W2554138451C98359873 @default.
• W2554138451 hasLocation W25541384511 @default.
• W2554138451 hasOpenAccess W2554138451 @default.
• W2554138451 hasPrimaryLocation W25541384511 @default.
• W2554138451 hasRelatedWork W116783880 @default.
• W2554138451 hasRelatedWork W1554947342 @default.
• W2554138451 hasRelatedWork W1647984221 @default.
• W2554138451 hasRelatedWork W2019961600 @default.
• W2554138451 hasRelatedWork W2087235292 @default.
• W2554138451 hasRelatedWork W2097157458 @default.
• W2554138451 hasRelatedWork W2133796939 @default.
• W2554138451 hasRelatedWork W2148991744 @default.
• W2554138451 hasRelatedWork W2181606736 @default.
• W2554138451 hasRelatedWork W2356455122 @default.
• W2554138451 hasRelatedWork W2364873297 @default.
• W2554138451 hasRelatedWork W2366287411 @default.
• W2554138451 hasRelatedWork W2405196115 @default.
• W2554138451 hasRelatedWork W2553912377 @default.
• W2554138451 hasRelatedWork W2744719763 @default.
• W2554138451 hasRelatedWork W2745764710 @default.
• W2554138451 hasRelatedWork W2786746258 @default.
• W2554138451 hasRelatedWork W2799069271 @default.
• W2554138451 hasRelatedWork W3096443866 @default.
• W2554138451 hasRelatedWork W2184894869 @default.
• W2554138451 isParatext "false" @default.
• W2554138451 isRetracted "false" @default.
• W2554138451 magId "2554138451" @default.
• W2554138451 workType "article" @default.