EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM
This paper presents the backpropagation algorithm based on an extended network approach in which the algorithm reduces to a graph labeling problem. This method is not only more general than the usual analytical derivations, which handle only the case of special network topologies, but also much easi...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
University of Petrosani
2012-12-01
|
| Series: | Annals of the University of Petrosani: Economics |
| Subjects: | |
| Online Access: | http://www.upet.ro/annals/economics/pdf/2012/part4/Petrini-1.pdf |
| _version_ | 1819053129304375296 |
|---|---|
| author | MIRCEA PETRINI |
| author_facet | MIRCEA PETRINI |
| author_sort | MIRCEA PETRINI |
| collection | DOAJ |
| description | This paper presents the backpropagation algorithm based on an extended network approach in which the algorithm reduces to a graph labeling problem. This method is not only more general than the usual analytical derivations, which handle only the case of special network topologies, but also much easier to follow. It also shows how the algorithm can be efficiently implemented in computing systems in which only local information can be transported through the network. |
| first_indexed | 2024-12-21T12:30:49Z |
| format | Article |
| id | doaj.art-e8d321906e0f48488e546b9435628dea |
| institution | Directory Open Access Journal |
| issn | 1582-5949 2247-8620 |
| language | English |
| last_indexed | 2024-12-21T12:30:49Z |
| publishDate | 2012-12-01 |
| publisher | University of Petrosani |
| record_format | Article |
| series | Annals of the University of Petrosani: Economics |
| spelling | doaj.art-e8d321906e0f48488e546b9435628dea2022-12-21T19:04:03ZengUniversity of PetrosaniAnnals of the University of Petrosani: Economics1582-59492247-86202012-12-01XII4177184EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHMMIRCEA PETRINI0University of Petrosani, RomaniaThis paper presents the backpropagation algorithm based on an extended network approach in which the algorithm reduces to a graph labeling problem. This method is not only more general than the usual analytical derivations, which handle only the case of special network topologies, but also much easier to follow. It also shows how the algorithm can be efficiently implemented in computing systems in which only local information can be transported through the network.http://www.upet.ro/annals/economics/pdf/2012/part4/Petrini-1.pdfArtificial Neural Network (ANN)backpropagationextended networkfeed-forward computationtraining pattern |
| spellingShingle | MIRCEA PETRINI EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM Annals of the University of Petrosani: Economics Artificial Neural Network (ANN) backpropagation extended network feed-forward computation training pattern |
| title | EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM |
| title_full | EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM |
| title_fullStr | EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM |
| title_full_unstemmed | EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM |
| title_short | EXTENDED NETWORK FOR BACKPROPAGATION ALGORITHM |
| title_sort | extended network for backpropagation algorithm |
| topic | Artificial Neural Network (ANN) backpropagation extended network feed-forward computation training pattern |
| url | http://www.upet.ro/annals/economics/pdf/2012/part4/Petrini-1.pdf |
| work_keys_str_mv | AT mirceapetrini extendednetworkforbackpropagationalgorithm |