<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<site>plutao.sid.inpe.br 800</site>
		<holdercode>{isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}</holdercode>
		<identifier>J8LNKAN8RW/3D53LKS</identifier>
		<repository>dpi.inpe.br/plutao/2012/11.28.16.47.57</repository>
		<lastupdate>2015:03.16.17.03.40 dpi.inpe.br/plutao@80/2008/08.19.15.01 administrator</lastupdate>
		<metadatarepository>dpi.inpe.br/plutao/2012/11.28.16.47.58</metadatarepository>
		<metadatalastupdate>2018:06.05.00.02.06 dpi.inpe.br/plutao@80/2008/08.19.15.01 administrator {D 2012}</metadatalastupdate>
		<secondarykey>INPE--PRE/</secondarykey>
		<label>lattes: 2720072834057575 2 SambattiAnLuCaShCa:2012:MPMeAu</label>
		<citationkey>SambattiAnLuCaShCa:2012:MPMeAu</citationkey>
		<title>MPCA Meta-Heuristics for automatic architecture optimization of a supervised artificial neural network</title>
		<year>2012</year>
		<secondarytype>PRE CI</secondarytype>
		<numberoffiles>1</numberoffiles>
		<size>147 KiB</size>
		<author>Sambatti, Sabrina Bergoch Monteiro,</author>
		<author>Anochi, Juliana Aparecida,</author>
		<author>Luz, Eduardo F. Pacheco da,</author>
		<author>Carvalho, Adenilson R.,</author>
		<author>Shiguemori, Elcio Hideiti,</author>
		<author>Campos Velho, Haroldo Fraga de,</author>
		<resumeid></resumeid>
		<resumeid></resumeid>
		<resumeid></resumeid>
		<resumeid></resumeid>
		<resumeid></resumeid>
		<resumeid>8JMKD3MGP5W/3C9JHC3</resumeid>
		<group>LAC-CTE-INPE-MCTI-GOV-BR</group>
		<group>LAC-CTE-INPE-MCTI-GOV-BR</group>
		<group></group>
		<group></group>
		<group></group>
		<group>LAC-CTE-INPE-MCTI-GOV-BR</group>
		<affiliation>Instituto Nacional de Pesquisas Espaciais (INPE)</affiliation>
		<affiliation>Instituto Nacional de Pesquisas Espaciais (INPE)</affiliation>
		<affiliation></affiliation>
		<affiliation>SERPRO</affiliation>
		<affiliation>Instituto de Estudos Avançado (IEAv)</affiliation>
		<affiliation>Instituto Nacional de Pesquisas Espaciais (INPE)</affiliation>
		<electronicmailaddress>sabrinabms@gmail.com</electronicmailaddress>
		<electronicmailaddress>juliana.anochi@lac.inpe.br</electronicmailaddress>
		<electronicmailaddress></electronicmailaddress>
		<electronicmailaddress></electronicmailaddress>
		<electronicmailaddress></electronicmailaddress>
		<electronicmailaddress>haroldo@lac.inpe.br</electronicmailaddress>
		<e-mailaddress>juliana.anochi@lac.inpe.br</e-mailaddress>
		<conferencename>World Congress on Computational Mechanics, 10 (WCCM).</conferencename>
		<conferencelocation>São Paulo</conferencelocation>
		<date>2012</date>
		<booktitle>Proceedings</booktitle>
		<tertiarytype>Paper</tertiarytype>
		<transferableflag>1</transferableflag>
		<contenttype>External Contribution</contenttype>
		<versiontype>finaldraft</versiontype>
		<abstract>Artificial neural networks (ANN) has been studied intensively, but there still are many unresolved issues. The search and definition of an optimal architecture remains a very relevant ANN research topic. The search space of neural network topology, each point represents a possible architecture. Associating each point to a performance level relies on the a priori establishment of some optimality criterion. Here, a new meta-heuristics, multi-particle collision algorithm (MPCA) was applied to design an optimum architecture for a supervised ANN. The MPCA optimization algorithm emulates a collision process of multiple particles inspired in processes of a neutron traveling in a nuclear reactor. The multilayer perceptron (MLP) was the neural network adopted here, and backpropagation strategy was used for calculating of the weight of connections to the MLP-NN. The MLP-NN configured by this optimal or inverse designs was applied to predict the seasonal mesoscale climate. The dataset for trainning is obtained from NCEP-NOAA reanalysis and from a metherological model. In order to reduce the dimension of the search space to find the optimized ANN, it is considered the following: three activation functions, up to three hidden layers, and up to 32 neurons per hidden layer. The comparison is performed between the ANN configuration obtained by automatic process and another configuration proposed by a human specialist.</abstract>
		<area>COMP</area>
		<language>en</language>
		<targetfile>sambatti_mpca.pdf</targetfile>
		<usergroup>lattes</usergroup>
		<usergroup>marciana</usergroup>
		<readergroup>administrator</readergroup>
		<readergroup>marciana</readergroup>
		<visibility>shown</visibility>
		<readpermission>allow from all</readpermission>
		<documentstage>not transferred</documentstage>
		<nexthigherunit>8JMKD3MGPCW/3ESGTTP</nexthigherunit>
		<citingitemlist>sid.inpe.br/mtc-m21/2012/07.13.14.49.40 3</citingitemlist>
		<hostcollection>dpi.inpe.br/plutao@80/2008/08.19.15.01</hostcollection>
		<username>marciana</username>
		<lasthostcollection>dpi.inpe.br/plutao@80/2008/08.19.15.01</lasthostcollection>
		<url>http://plutao.sid.inpe.br/rep-/dpi.inpe.br/plutao/2012/11.28.16.47.57</url>
	</metadata>
</metadatalist>