Mostra el registre d'ítem simple
Convolutional neural network training with dynamic epoch ordering
dc.contributor.author | Plana Rius, Ferran |
dc.contributor.author | Angulo Bahón, Cecilio |
dc.contributor.author | Casas, Marc |
dc.contributor.author | Mirats Tur, Josep Maria |
dc.contributor.other | Universitat Politècnica de Catalunya. Doctorat en Automàtica, Robòtica i Visió |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament d'Enginyeria de Sistemes, Automàtica i Informàtica Industrial |
dc.contributor.other | Barcelona Supercomputing Center |
dc.date.accessioned | 2020-02-10T10:26:55Z |
dc.date.issued | 2019 |
dc.identifier.citation | Plana, F. [et al.]. Convolutional neural network training with dynamic epoch ordering. A: International Conference of the Catalan Association for Artificial Intelligence. "Artificial Intelligence Research and Development vol. 319". IOS Press, 2019, p. 105-114. |
dc.identifier.isbn | 978-1-64368-014-9 |
dc.identifier.uri | http://hdl.handle.net/2117/177282 |
dc.description.abstract | The paper presented exposes a novel approach to feed data to a Convolutional Neural Network (CNN) while training. Normally, neural networks are fed with shuffled data without any control of what type of examples contains a minibatch. For situations where data are abundant and there does not exist an unbalancing between classes, shuffling the training data is enough to ensure a balanced mini-batch. On the contrary, most real-world problems end up with databases where some classes are predominant vs others, ill-conditioning the training network to learn those classes forgetting the others. For those conditioned cases, most common methods simply discard a certain number of samples until the data is balanced, but this paper proposes an ordered method of feeding data while preserving randomness in the mini-batch composition and using all available samples. This method has proven to solve the problem with unbalanced data-sets while competing with other methods. Moreover, the paper will focus its attention to a well know CNN network structure, named Deep Residual Networks. |
dc.format.extent | 10 p. |
dc.language.iso | eng |
dc.publisher | IOS Press |
dc.rights | Attribution-NonCommercial-NoDerivs 3.0 Spain |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
dc.subject | Àrees temàtiques de la UPC::Informàtica |
dc.subject.lcsh | Neural networks (Computer science) |
dc.subject.lcsh | Artificial intelligence |
dc.subject.other | Deep learning |
dc.subject.other | Supervised learning |
dc.subject.other | Convolutional Neural Networks |
dc.subject.other | Data management |
dc.subject.other | Deep residual networks |
dc.title | Convolutional neural network training with dynamic epoch ordering |
dc.type | Conference lecture |
dc.subject.lemac | Xarxes neuronals (Informàtica) |
dc.subject.lemac | Intel·ligència artificial |
dc.contributor.group | Universitat Politècnica de Catalunya. GREC - Grup de Recerca en Enginyeria del Coneixement |
dc.identifier.doi | 10.3233/FAIA190113 |
dc.description.peerreviewed | Peer Reviewed |
dc.relation.publisherversion | http://ebooks.iospress.nl/volumearticle/52826 |
dc.rights.access | Restricted access - publisher's policy |
local.identifier.drac | 25906749 |
dc.description.version | Postprint (published version) |
dc.date.lift | 10000-01-01 |
local.citation.author | Plana, F.; Angulo, C.; Casas, M.; Mirats, J. |
local.citation.contributor | International Conference of the Catalan Association for Artificial Intelligence |
local.citation.publicationName | Artificial Intelligence Research and Development vol. 319 |
local.citation.startingPage | 105 |
local.citation.endingPage | 114 |