The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Este artigo apresenta um algoritmo eficiente para tarefas de aprendizagem multissistema em larga escala. A arquitetura proposta, denominada 'RBF×SOM', é baseada no SOM2, isto é, um'SOM de SOMs'. Como é o caso da rede modular SOM (mnSOM) com módulos perceptron multicamadas (MLP-mnSOM), o objetivo do RBF×SOM é organizar um mapa contínuo de funções não lineares representando relações de entrada-saída multiclasse dos conjuntos de dados fornecidos. . Ao adotar o algoritmo para o SOM2, o RBF×SOM gera um mapa muito mais rápido que o mnSOM original e sem o problema de mínimos locais. Além disso, o RBF×SOM pode ser aplicado a casos mais difíceis, que não foram facilmente resolvidos pelo MLP-mnSOM. Assim, o RBF×SOM pode lidar com casos em que a densidade de probabilidade das entradas é dependente das classes. Isso tende a acontecer com mais frequência à medida que a dimensão de entrada aumenta. O RBF×SOM, portanto, supera muitos dos problemas inerentes ao MLP-mnSOM, e isso é crucial para aplicação em tarefas de grande escala. Os resultados da simulação com conjuntos de dados artificiais e um conjunto de dados meteorológicos confirmam o desempenho do RBF×SOM.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Takashi OHKUBO, Kazuhiro TOKUNAGA, Tetsuo FURUKAWA, "RBFSOM: An Efficient Algorithm for Large-Scale Multi-System Learning" in IEICE TRANSACTIONS on Information,
vol. E92-D, no. 7, pp. 1388-1396, July 2009, doi: 10.1587/transinf.E92.D.1388.
Abstract: This paper presents an efficient algorithm for large-scale multi-system learning task. The proposed architecture, referred to as the 'RBF×SOM', is based on the SOM2, that is, a'SOM of SOMs'. As is the case in the modular network SOM (mnSOM) with multilayer perceptron modules (MLP-mnSOM), the aim of the RBF×SOM is to organize a continuous map of nonlinear functions representing multi-class input-output relations of the given datasets. By adopting the algorithm for the SOM2, the RBF×SOM generates a map much faster than the original mnSOM, and without the local minima problem. In addition, the RBF×SOM can be applied to more difficult cases, that were not easily dealt with by the MLP-mnSOM. Thus, the RBF×SOM can deal with cases in which the probability density of the inputs is dependent on the classes. This tends to happen more often as the input dimension increases. The RBF×SOM therefore, overcomes many of the problems inherent in the MLP-mnSOM, and this is crucial for application to large scale tasks. Simulation results with artificial datasets and a meteorological dataset confirm the performance of the RBF×SOM.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E92.D.1388/_p
Copiar
@ARTICLE{e92-d_7_1388,
author={Takashi OHKUBO, Kazuhiro TOKUNAGA, Tetsuo FURUKAWA, },
journal={IEICE TRANSACTIONS on Information},
title={RBFSOM: An Efficient Algorithm for Large-Scale Multi-System Learning},
year={2009},
volume={E92-D},
number={7},
pages={1388-1396},
abstract={This paper presents an efficient algorithm for large-scale multi-system learning task. The proposed architecture, referred to as the 'RBF×SOM', is based on the SOM2, that is, a'SOM of SOMs'. As is the case in the modular network SOM (mnSOM) with multilayer perceptron modules (MLP-mnSOM), the aim of the RBF×SOM is to organize a continuous map of nonlinear functions representing multi-class input-output relations of the given datasets. By adopting the algorithm for the SOM2, the RBF×SOM generates a map much faster than the original mnSOM, and without the local minima problem. In addition, the RBF×SOM can be applied to more difficult cases, that were not easily dealt with by the MLP-mnSOM. Thus, the RBF×SOM can deal with cases in which the probability density of the inputs is dependent on the classes. This tends to happen more often as the input dimension increases. The RBF×SOM therefore, overcomes many of the problems inherent in the MLP-mnSOM, and this is crucial for application to large scale tasks. Simulation results with artificial datasets and a meteorological dataset confirm the performance of the RBF×SOM.},
keywords={},
doi={10.1587/transinf.E92.D.1388},
ISSN={1745-1361},
month={July},}
Copiar
TY - JOUR
TI - RBFSOM: An Efficient Algorithm for Large-Scale Multi-System Learning
T2 - IEICE TRANSACTIONS on Information
SP - 1388
EP - 1396
AU - Takashi OHKUBO
AU - Kazuhiro TOKUNAGA
AU - Tetsuo FURUKAWA
PY - 2009
DO - 10.1587/transinf.E92.D.1388
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E92-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2009
AB - This paper presents an efficient algorithm for large-scale multi-system learning task. The proposed architecture, referred to as the 'RBF×SOM', is based on the SOM2, that is, a'SOM of SOMs'. As is the case in the modular network SOM (mnSOM) with multilayer perceptron modules (MLP-mnSOM), the aim of the RBF×SOM is to organize a continuous map of nonlinear functions representing multi-class input-output relations of the given datasets. By adopting the algorithm for the SOM2, the RBF×SOM generates a map much faster than the original mnSOM, and without the local minima problem. In addition, the RBF×SOM can be applied to more difficult cases, that were not easily dealt with by the MLP-mnSOM. Thus, the RBF×SOM can deal with cases in which the probability density of the inputs is dependent on the classes. This tends to happen more often as the input dimension increases. The RBF×SOM therefore, overcomes many of the problems inherent in the MLP-mnSOM, and this is crucial for application to large scale tasks. Simulation results with artificial datasets and a meteorological dataset confirm the performance of the RBF×SOM.
ER -