On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures
- Author
- BIANCHINI, Monica1 ; SCARSELLI, Franco1
[1] University of Siena, Siena 53100, Italy - Source
IEEE transactions on neural networks and learning systems (Print). 2014, Vol 25, Num 8, pp 1553-1565, 13 p ; ref : 54 ref
- ISSN
- 2162-237X
- Scientific domain
- Cognition; Electronics; Computer science; Psychology, psychopathology, psychiatry
- Publisher
- Institute of Electrical and Electronics Engineers, New York, NY
- Publication country
- United States
- Document type
- Article
- Language
- English
- Author keyword
- Betti numbers Vapnik-Chervonenkis dimension (VC-dim) deep neural networks function approximation topological complexity
- Keyword (fr)
- Apprentissage probabilités Borne inférieure Borne supérieure Boucle anticipation Classification automatique Fonction activation Grande profondeur Méthode heuristique Réseau neuronal Résultat expérimental Topologie Transformation sigmoïdale Apprentissage profond Approximation d'une fonction Réseau neuronal non bouclé
- Keyword (en)
- Probability learning Lower bound Upper bound Feedforward Automatic classification Activation function Great depth Heuristic method Neural network Experimental result Topology Sigmoidal transformation Deep learning Function approximation Feedforward neural nets
- Keyword (es)
- Aprendizaje probabilidades Cota inferior Cota superior Ciclo anticipación Clasificación automática Función actividad Gran profundidad Método heurístico Red neuronal Resultado experimental Topología transformación sigmoidal Aprendizaje profundo Aproximación de funciones Red neural unidireccional
- Classification
- Pascal
- 001 Exact sciences and technology / 001D Applied sciences / 001D02 Computer science; control theory; systems / 001D02B Software / 001D02B07 Memory organisation. Data processing / 001D02B07B Data processing. List processing. Character string processing
- Pascal
- 001 Exact sciences and technology / 001D Applied sciences / 001D02 Computer science; control theory; systems / 001D02C Artificial intelligence / 001D02C03 Pattern recognition. Digital image processing. Computational geometry
- Pascal
- 001 Exact sciences and technology / 001D Applied sciences / 001D02 Computer science; control theory; systems / 001D02C Artificial intelligence / 001D02C06 Connectionism. Neural networks
- Discipline
- Computer science : theoretical automation and systems
- Origin
- Inist-CNRS
- Database
- PASCAL
- INIST identifier
- 28747107
Sauf mention contraire ci-dessus, le contenu de cette notice bibliographique peut être utilisé dans le cadre d’une licence CC BY 4.0 Inist-CNRS / Unless otherwise stated above, the content of this bibliographic record may be used under a CC BY 4.0 licence by Inist-CNRS / A menos que se haya señalado antes, el contenido de este registro bibliográfico puede ser utilizado al amparo de una licencia CC BY 4.0 Inist-CNRS
Access to the document
Install on your browser
DOI
10.1109/TNNLS.2013.2293637 
https://dx.doi.org/10.1109/TNNLS.2013.2293637
"O:13:\"PanistOpenUrl\":36:{s:10:\"\u0000*\u0000openUrl\";N;s:6:\"\u0000*\u0000idc\";N;s:6:\"\u0000*\u0000fmt\";s:7:\"journal\";s:6:\"\u0000*\u0000doi\";s:26:\"10.1109\/TNNLS.2013.2293637\";s:6:\"\u0000*\u0000pii\";s:0:\"\";s:7:\"\u0000*\u0000pmid\";s:0:\"\";s:9:\"\u0000*\u0000atitle\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:9:\"\u0000*\u0000jtitle\";s:65:\"IEEE transactions on neural networks and learning systems (Print)\";s:9:\"\u0000*\u0000stitle\";s:45:\"IEEE trans. neural netw. learn. syst. (Print)\";s:7:\"\u0000*\u0000date\";s:4:\"2014\";s:9:\"\u0000*\u0000volume\";s:2:\"25\";s:8:\"\u0000*\u0000issue\";s:1:\"8\";s:8:\"\u0000*\u0000spage\";s:4:\"1553\";s:8:\"\u0000*\u0000epage\";s:4:\"1565\";s:8:\"\u0000*\u0000pages\";s:0:\"\";s:7:\"\u0000*\u0000issn\";s:9:\"2162-237X\";s:8:\"\u0000*\u0000eissn\";s:0:\"\";s:9:\"\u0000*\u0000aulast\";s:9:\"BIANCHINI\";s:10:\"\u0000*\u0000aufirst\";s:6:\"Monica\";s:9:\"\u0000*\u0000auinit\";s:0:\"\";s:10:\"\u0000*\u0000auinitm\";s:0:\"\";s:5:\"\u0000*\u0000au\";a:1:{i:0;s:17:\"SCARSELLI, Franco\";}s:9:\"\u0000*\u0000aucorp\";s:0:\"\";s:7:\"\u0000*\u0000isbn\";s:0:\"\";s:8:\"\u0000*\u0000coden\";s:0:\"\";s:8:\"\u0000*\u0000genre\";s:7:\"article\";s:7:\"\u0000*\u0000part\";s:0:\"\";s:9:\"\u0000*\u0000btitle\";s:0:\"\";s:8:\"\u0000*\u0000title\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:8:\"\u0000*\u0000place\";s:12:\"New York, NY\";s:6:\"\u0000*\u0000pub\";s:49:\"Institute of Electrical and Electronics Engineers\";s:10:\"\u0000*\u0000edition\";s:0:\"\";s:9:\"\u0000*\u0000tpages\";s:2:\"13\";s:9:\"\u0000*\u0000series\";s:0:\"\";s:8:\"\u0000*\u0000proxy\";s:30:\"http:\/\/proxyout.inist.fr:8080\/\";s:12:\"\u0000*\u0000integrite\";b:1;}"
Panist access
"O:12:\"IstexOpenUrl\":35:{s:10:\"\u0000*\u0000openUrl\";N;s:6:\"\u0000*\u0000fmt\";s:7:\"journal\";s:6:\"\u0000*\u0000doi\";s:26:\"10.1109\/TNNLS.2013.2293637\";s:6:\"\u0000*\u0000pii\";s:0:\"\";s:7:\"\u0000*\u0000pmid\";s:0:\"\";s:9:\"\u0000*\u0000atitle\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:9:\"\u0000*\u0000jtitle\";s:65:\"IEEE transactions on neural networks and learning systems (Print)\";s:9:\"\u0000*\u0000stitle\";s:45:\"IEEE trans. neural netw. learn. syst. (Print)\";s:7:\"\u0000*\u0000date\";s:4:\"2014\";s:9:\"\u0000*\u0000volume\";s:2:\"25\";s:8:\"\u0000*\u0000issue\";s:1:\"8\";s:8:\"\u0000*\u0000spage\";s:4:\"1553\";s:8:\"\u0000*\u0000epage\";s:4:\"1565\";s:8:\"\u0000*\u0000pages\";s:0:\"\";s:7:\"\u0000*\u0000issn\";s:9:\"2162-237X\";s:8:\"\u0000*\u0000eissn\";s:0:\"\";s:9:\"\u0000*\u0000aulast\";s:9:\"BIANCHINI\";s:10:\"\u0000*\u0000aufirst\";s:6:\"Monica\";s:9:\"\u0000*\u0000auinit\";s:0:\"\";s:10:\"\u0000*\u0000auinitm\";s:0:\"\";s:5:\"\u0000*\u0000au\";a:1:{i:0;s:17:\"SCARSELLI, Franco\";}s:9:\"\u0000*\u0000aucorp\";s:0:\"\";s:7:\"\u0000*\u0000isbn\";s:0:\"\";s:8:\"\u0000*\u0000coden\";s:0:\"\";s:8:\"\u0000*\u0000genre\";s:7:\"article\";s:7:\"\u0000*\u0000part\";s:0:\"\";s:9:\"\u0000*\u0000btitle\";s:0:\"\";s:8:\"\u0000*\u0000title\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:8:\"\u0000*\u0000place\";s:12:\"New York, NY\";s:6:\"\u0000*\u0000pub\";s:49:\"Institute of Electrical and Electronics Engineers\";s:10:\"\u0000*\u0000edition\";s:0:\"\";s:9:\"\u0000*\u0000tpages\";s:2:\"13\";s:9:\"\u0000*\u0000series\";s:0:\"\";s:8:\"\u0000*\u0000proxy\";s:30:\"http:\/\/proxyout.inist.fr:8080\/\";s:12:\"\u0000*\u0000integrite\";b:1;}"
Istex access
"O:16:\"UnpaywallOpenUrl\":35:{s:8:\"\u0000*\u0000email\";s:27:\"assistance-portail@inist.fr\";s:6:\"\u0000*\u0000fmt\";N;s:6:\"\u0000*\u0000doi\";s:26:\"10.1109\/TNNLS.2013.2293637\";s:6:\"\u0000*\u0000pii\";N;s:7:\"\u0000*\u0000pmid\";N;s:9:\"\u0000*\u0000atitle\";N;s:9:\"\u0000*\u0000jtitle\";N;s:9:\"\u0000*\u0000stitle\";N;s:7:\"\u0000*\u0000date\";N;s:9:\"\u0000*\u0000volume\";N;s:8:\"\u0000*\u0000issue\";N;s:8:\"\u0000*\u0000spage\";N;s:8:\"\u0000*\u0000epage\";N;s:8:\"\u0000*\u0000pages\";N;s:7:\"\u0000*\u0000issn\";N;s:8:\"\u0000*\u0000eissn\";N;s:9:\"\u0000*\u0000aulast\";N;s:10:\"\u0000*\u0000aufirst\";N;s:9:\"\u0000*\u0000auinit\";N;s:10:\"\u0000*\u0000auinitm\";N;s:5:\"\u0000*\u0000au\";N;s:9:\"\u0000*\u0000aucorp\";N;s:7:\"\u0000*\u0000isbn\";N;s:8:\"\u0000*\u0000coden\";N;s:8:\"\u0000*\u0000genre\";N;s:7:\"\u0000*\u0000part\";N;s:9:\"\u0000*\u0000btitle\";N;s:8:\"\u0000*\u0000title\";N;s:8:\"\u0000*\u0000place\";N;s:6:\"\u0000*\u0000pub\";N;s:10:\"\u0000*\u0000edition\";N;s:9:\"\u0000*\u0000tpages\";N;s:9:\"\u0000*\u0000series\";N;s:8:\"\u0000*\u0000proxy\";s:30:\"http:\/\/proxyout.inist.fr:8080\/\";s:12:\"\u0000*\u0000integrite\";N;}"
Unpaywall access
"O:16:\"EuropePMCOpenUrl\":35:{s:10:\"\u0000*\u0000openUrl\";N;s:6:\"\u0000*\u0000fmt\";s:7:\"journal\";s:6:\"\u0000*\u0000doi\";s:26:\"10.1109\/TNNLS.2013.2293637\";s:6:\"\u0000*\u0000pii\";s:0:\"\";s:7:\"\u0000*\u0000pmid\";s:0:\"\";s:9:\"\u0000*\u0000atitle\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:9:\"\u0000*\u0000jtitle\";s:65:\"IEEE transactions on neural networks and learning systems (Print)\";s:9:\"\u0000*\u0000stitle\";s:45:\"IEEE trans. neural netw. learn. syst. (Print)\";s:7:\"\u0000*\u0000date\";s:4:\"2014\";s:9:\"\u0000*\u0000volume\";s:2:\"25\";s:8:\"\u0000*\u0000issue\";s:1:\"8\";s:8:\"\u0000*\u0000spage\";s:4:\"1553\";s:8:\"\u0000*\u0000epage\";s:4:\"1565\";s:8:\"\u0000*\u0000pages\";s:0:\"\";s:7:\"\u0000*\u0000issn\";s:9:\"2162-237X\";s:8:\"\u0000*\u0000eissn\";s:0:\"\";s:9:\"\u0000*\u0000aulast\";s:9:\"BIANCHINI\";s:10:\"\u0000*\u0000aufirst\";s:6:\"Monica\";s:9:\"\u0000*\u0000auinit\";s:0:\"\";s:10:\"\u0000*\u0000auinitm\";s:0:\"\";s:5:\"\u0000*\u0000au\";a:1:{i:0;s:17:\"SCARSELLI, Franco\";}s:9:\"\u0000*\u0000aucorp\";s:0:\"\";s:7:\"\u0000*\u0000isbn\";s:0:\"\";s:8:\"\u0000*\u0000coden\";s:0:\"\";s:8:\"\u0000*\u0000genre\";s:7:\"article\";s:7:\"\u0000*\u0000part\";s:0:\"\";s:9:\"\u0000*\u0000btitle\";s:0:\"\";s:8:\"\u0000*\u0000title\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:8:\"\u0000*\u0000place\";s:12:\"New York, NY\";s:6:\"\u0000*\u0000pub\";s:49:\"Institute of Electrical and Electronics Engineers\";s:10:\"\u0000*\u0000edition\";s:0:\"\";s:9:\"\u0000*\u0000tpages\";s:2:\"13\";s:9:\"\u0000*\u0000series\";s:0:\"\";s:8:\"\u0000*\u0000proxy\";s:30:\"http:\/\/proxyout.inist.fr:8080\/\";s:12:\"\u0000*\u0000integrite\";b:1;}"
Europe PubMed Central access
"O:11:\"BaseOpenUrl\":35:{s:10:\"\u0000*\u0000openUrl\";N;s:6:\"\u0000*\u0000fmt\";s:7:\"journal\";s:6:\"\u0000*\u0000doi\";s:26:\"10.1109\/TNNLS.2013.2293637\";s:6:\"\u0000*\u0000pii\";s:0:\"\";s:7:\"\u0000*\u0000pmid\";s:0:\"\";s:9:\"\u0000*\u0000atitle\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:9:\"\u0000*\u0000jtitle\";s:65:\"IEEE transactions on neural networks and learning systems (Print)\";s:9:\"\u0000*\u0000stitle\";s:45:\"IEEE trans. neural netw. learn. syst. (Print)\";s:7:\"\u0000*\u0000date\";s:4:\"2014\";s:9:\"\u0000*\u0000volume\";s:2:\"25\";s:8:\"\u0000*\u0000issue\";s:1:\"8\";s:8:\"\u0000*\u0000spage\";s:4:\"1553\";s:8:\"\u0000*\u0000epage\";s:4:\"1565\";s:8:\"\u0000*\u0000pages\";s:0:\"\";s:7:\"\u0000*\u0000issn\";s:9:\"2162-237X\";s:8:\"\u0000*\u0000eissn\";s:0:\"\";s:9:\"\u0000*\u0000aulast\";s:9:\"BIANCHINI\";s:10:\"\u0000*\u0000aufirst\";s:6:\"Monica\";s:9:\"\u0000*\u0000auinit\";s:0:\"\";s:10:\"\u0000*\u0000auinitm\";s:0:\"\";s:5:\"\u0000*\u0000au\";a:1:{i:0;s:17:\"SCARSELLI, Franco\";}s:9:\"\u0000*\u0000aucorp\";s:0:\"\";s:7:\"\u0000*\u0000isbn\";s:0:\"\";s:8:\"\u0000*\u0000coden\";s:0:\"\";s:8:\"\u0000*\u0000genre\";s:7:\"article\";s:7:\"\u0000*\u0000part\";s:0:\"\";s:9:\"\u0000*\u0000btitle\";s:0:\"\";s:8:\"\u0000*\u0000title\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:8:\"\u0000*\u0000place\";s:12:\"New York, NY\";s:6:\"\u0000*\u0000pub\";s:49:\"Institute of Electrical and Electronics Engineers\";s:10:\"\u0000*\u0000edition\";s:0:\"\";s:9:\"\u0000*\u0000tpages\";s:2:\"13\";s:9:\"\u0000*\u0000series\";s:0:\"\";s:8:\"\u0000*\u0000proxy\";s:30:\"http:\/\/proxyout.inist.fr:8080\/\";s:12:\"\u0000*\u0000integrite\";b:1;}"
Base Search access
"O:12:\"ArXivOpenUrl\":35:{s:10:\"\u0000*\u0000openUrl\";N;s:6:\"\u0000*\u0000fmt\";s:7:\"journal\";s:6:\"\u0000*\u0000doi\";s:26:\"10.1109\/TNNLS.2013.2293637\";s:6:\"\u0000*\u0000pii\";s:0:\"\";s:7:\"\u0000*\u0000pmid\";s:0:\"\";s:9:\"\u0000*\u0000atitle\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:9:\"\u0000*\u0000jtitle\";s:65:\"IEEE transactions on neural networks and learning systems (Print)\";s:9:\"\u0000*\u0000stitle\";s:45:\"IEEE trans. neural netw. learn. syst. (Print)\";s:7:\"\u0000*\u0000date\";s:4:\"2014\";s:9:\"\u0000*\u0000volume\";s:2:\"25\";s:8:\"\u0000*\u0000issue\";s:1:\"8\";s:8:\"\u0000*\u0000spage\";s:4:\"1553\";s:8:\"\u0000*\u0000epage\";s:4:\"1565\";s:8:\"\u0000*\u0000pages\";s:0:\"\";s:7:\"\u0000*\u0000issn\";s:9:\"2162-237X\";s:8:\"\u0000*\u0000eissn\";s:0:\"\";s:9:\"\u0000*\u0000aulast\";s:9:\"BIANCHINI\";s:10:\"\u0000*\u0000aufirst\";s:6:\"Monica\";s:9:\"\u0000*\u0000auinit\";s:0:\"\";s:10:\"\u0000*\u0000auinitm\";s:0:\"\";s:5:\"\u0000*\u0000au\";a:1:{i:0;s:17:\"SCARSELLI, Franco\";}s:9:\"\u0000*\u0000aucorp\";s:0:\"\";s:7:\"\u0000*\u0000isbn\";s:0:\"\";s:8:\"\u0000*\u0000coden\";s:0:\"\";s:8:\"\u0000*\u0000genre\";s:7:\"article\";s:7:\"\u0000*\u0000part\";s:0:\"\";s:9:\"\u0000*\u0000btitle\";s:0:\"\";s:8:\"\u0000*\u0000title\";s:100:\"On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures\";s:8:\"\u0000*\u0000place\";s:12:\"New York, NY\";s:6:\"\u0000*\u0000pub\";s:49:\"Institute of Electrical and Electronics Engineers\";s:10:\"\u0000*\u0000edition\";s:0:\"\";s:9:\"\u0000*\u0000tpages\";s:2:\"13\";s:9:\"\u0000*\u0000series\";s:0:\"\";s:8:\"\u0000*\u0000proxy\";s:30:\"http:\/\/proxyout.inist.fr:8080\/\";s:12:\"\u0000*\u0000integrite\";b:1;}"
arXiv access
