|1/2013 - 1|
Automatic and Parallel Optimized Learning for Neural Networks performing MIMO ApplicationsFULGINEI, F. R. , LAUDANI, A. , SALVINI, A. , PARODI, M.
|Click to see author's profile on SCOPUS, IEEE Xplore, Web of Science|
|Download PDF (705 KB) | Citation | Downloads: 1,278 | Views: 4,289|
neural networks, multivariate function decomposition, learning optimization, parallel computing, genetic algorithms
neural(23), networks(14), network(9), salvini(6), riganti(6), fulginei(6), decomposition(5), problems(4), optimization(4), feed(4)
Blue keywords are present in both the references section and the paper title.
About this article
Date of Publication: 2013-02-28
Volume 13, Issue 1, Year 2013, On page(s): 3 - 12
ISSN: 1582-7445, e-ISSN: 1844-7600
Digital Object Identifier: 10.4316/AECE.2013.01001
Web of Science Accession Number: 000315768300001
SCOPUS ID: 84875323616
An automatic and optimized approach based on multivariate functions decomposition is presented to face Multi-Input-Multi-Output (MIMO) applications by using Single-Input-Single-Output (SISO) feed-forward Neural Networks (NNs). Indeed, often the learning time and the computational costs are too large for an effective use of MIMO NNs. Since performing a MISO neural model by starting from a single MIMO NN is frequently adopted in literature, the proposed method introduces three other steps: 1) a further decomposition; 2) a learning optimization; 3) a parallel training to speed up the process. Starting from a MISO NN, a collection of SISO NNs can be obtained by means a multi-dimensional Single Value Decomposition (SVD). Then, a general approach for the learning optimization of SISO NNs is applied. It is based on the observation that the performances of SISO NNs improve in terms of generalization and robustness against noise under suitable learning conditions. Thus, each SISO NN is trained and optimized by using limited training data that allow a significant decrease of computational costs. Moreover, a parallel architecture can be easily implemented. Consequently, the presented approach allows to perform an automatic conversion of MIMO NN into a collection of parallel-optimized SISO NNs. Experimental results will be suitably shown.
Web of Science® Times Cited: 21 [View]
View record in Web of Science® [View]
View Related Records® [View]
Updated 2 days, 5 hours ago
SCOPUS® Times Cited: 29
View record in SCOPUS® [Free preview]
 On Training Efficiency and Computational Costs of a Feed Forward Neural Network: A Review, Laudani, Antonino, Lozito, Gabriele Maria, Riganti Fulginei, Francesco, Salvini, Alessandro, Computational Intelligence and Neuroscience, ISSN 1687-5265, Issue , 2015.
Digital Object Identifier: 10.1155/2015/818243 [CrossRef]
 Two FPGA-Oriented High-Speed Irradiance Virtual Sensors for Photovoltaic Plants, Oliveri, Alberto, Cassottana, Luca, Laudani, Antonino, Riganti Fulginei, Francesco, Lozito, Gabriele Maria, Salvini, Alessandro, Storace, Marco, IEEE Transactions on Industrial Informatics, ISSN 1551-3203, Issue 1, Volume 13, 2017.
Digital Object Identifier: 10.1109/TII.2015.2462293 [CrossRef]
 A New Neural Predictor for ELF Magnetic Field Strength, Coco, Salvatore, Laudani, Antonino, Fulginei, Francesco Riganti, Salvini, Alessandro, IEEE Transactions on Magnetics, ISSN 0018-9464, Issue 2, Volume 50, 2014.
Digital Object Identifier: 10.1109/TMAG.2013.2283022 [CrossRef]
 A Combined Methodology of Adaptive Neuro-Fuzzy Inference System and Genetic Algorithm for Short-term Energy Forecasting, KAMPOUROPOULOS, K., ANDRADE, F., GARCIA, A., ROMERAL, L., Advances in Electrical and Computer Engineering, ISSN 1582-7445, Issue 1, Volume 14, 2014.
Digital Object Identifier: 10.4316/AECE.2014.01002 [CrossRef] [Full text]
 SVM versus MAP on Accelerometer Data to Distinguish among Locomotor Activities Executed at Different Speeds, Schmid, Maurizio, Riganti-Fulginei, Francesco, Bernabucci, Ivan, Laudani, Antonino, Bibbo, Daniele, Muscillo, Rossana, Salvini, Alessandro, Conforto, Silvia, Computational and Mathematical Methods in Medicine, ISSN 1748-670X, Issue , 2013.
Digital Object Identifier: 10.1155/2013/343084 [CrossRef]
 A Neural-FEM tool for the 2-D magnetic hysteresis modeling, Cardelli, E., Faba, A., Laudani, A., Lozito, G.M., Riganti Fulginei, F., Salvini, A., Physica B: Condensed Matter, ISSN 0921-4526, Issue , 2016.
Digital Object Identifier: 10.1016/j.physb.2015.12.006 [CrossRef]
 A Neural Network-Based Low-Cost Solar Irradiance Sensor, Mancilla-David, Fernando, Riganti-Fulginei, Francesco, Laudani, Antonino, Salvini, Alessandro, IEEE Transactions on Instrumentation and Measurement, ISSN 0018-9456, Issue 3, Volume 63, 2014.
Digital Object Identifier: 10.1109/TIM.2013.2282005 [CrossRef]
 Two-dimensional magnetic modeling of ferromagnetic materials by using a neural networks based hybrid approach, Cardelli, E., Faba, A., Laudani, A., Lozito, G.M., Riganti Fulginei, F., Salvini, A., Physica B: Condensed Matter, ISSN 0921-4526, Issue , 2016.
Digital Object Identifier: 10.1016/j.physb.2015.12.005 [CrossRef]
 Estimation of the Earth Resistance by Artificial Neural Network Model, Asimakopoulou, Fani E., Kontargyri, Vassiliki T., Tsekouras, George J., Gonos, Ioannis F., Stathopulos, Ioannis A., IEEE Transactions on Industry Applications, ISSN 0093-9994, Issue 6, Volume 51, 2015.
Digital Object Identifier: 10.1109/TIA.2015.2427114 [CrossRef]
 Hybrid Neural Network Approach Based Tool for the Modelling of Photovoltaic Panels, Laudani, Antonino, Lozito, Gabriele Maria, Riganti Fulginei, Francesco, Salvini, Alessandro, International Journal of Photoenergy, ISSN 1110-662X, Issue , 2015.
Digital Object Identifier: 10.1155/2015/413654 [CrossRef]
 A neural approach for the numerical modeling of two-dimensional magnetic hysteresis, Cardelli, E., Faba, A., Laudani, A., Riganti Fulginei, F., Salvini, A., Journal of Applied Physics, ISSN 0021-8979, Issue 17, Volume 117, 2015.
Digital Object Identifier: 10.1063/1.4916306 [CrossRef]
 A Neural Network Embedded System for Real-time Estimation of Muscle Forces, Lozito, Gabriele Maria, Schmid, Maurizio, Conforto, Silvia, Fulginei, Francesco Riganti, Bibbo, Daniele, Procedia Computer Science, ISSN 1877-0509, Issue , 2015.
Digital Object Identifier: 10.1016/j.procs.2015.05.196 [CrossRef]
 Computer Modeling of Nickel–Iron Alloy in Power Electronics Applications, Cardelli, Ermanno, Faba, Antonio, Laudani, Antonino, Quondam Antonio, Simone, Riganti Fulginei, Francesco, Salvini, Alessandro, IEEE Transactions on Industrial Electronics, ISSN 0278-0046, Issue 3, Volume 64, 2017.
Digital Object Identifier: 10.1109/TIE.2016.2597129 [CrossRef]
Disclaimer: All information displayed above was retrieved by using remote connections to respective databases. For the best user experience, we update all data by using background processes, and use caches in order to reduce the load on the servers we retrieve the information from. As we have no control on the availability of the database servers and sometimes the Internet connectivity may be affected, we do not guarantee the information is correct or complete. For the most accurate data, please always consult the database sites directly. Some external links require authentication or an institutional subscription.
Web of Science® is a registered trademark of Thomson Reuters, Scopus® is a registered trademark of Elsevier B.V., other product names, company names, brand names, trademarks and logos are the property of their respective owners.
Faculty of Electrical Engineering and Computer Science
Stefan cel Mare University of Suceava, Romania
All rights reserved: Advances in Electrical and Computer Engineering is a registered trademark of the Stefan cel Mare University of Suceava. No part of this publication may be reproduced, stored in a retrieval system, photocopied, recorded or archived, without the written permission from the Editor. When authors submit their papers for publication, they agree that the copyright for their article be transferred to the Faculty of Electrical Engineering and Computer Science, Stefan cel Mare University of Suceava, Romania, if and only if the articles are accepted for publication. The copyright covers the exclusive rights to reproduce and distribute the article, including reprints and translations.
Permission for other use: The copyright owner's consent does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific written permission must be obtained from the Editor for such copying. Direct linking to files hosted on this website is strictly prohibited.
Disclaimer: Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinions or statements appear in this journal, they wish to make it clear that all information and opinions formulated in the articles, as well as linguistic accuracy, are the sole responsibility of the author.