Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 640499

Parallel levenberg-marquardt-based neural network with variable decay rate


Baček, Tomislav; Majetić, Dubravko; Brezak, Danko
Parallel levenberg-marquardt-based neural network with variable decay rate // CIM 2013 : Computer integrated manufacturing and high speed machining / Abele, Eberhard ; Udiljak, Toma ; Ciglar, Damir (ur.).
Zagreb: Croatian Association of Production Engineering, 2013. str. 59-64 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)


CROSBI ID: 640499 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
Parallel levenberg-marquardt-based neural network with variable decay rate

Autori
Baček, Tomislav ; Majetić, Dubravko ; Brezak, Danko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
CIM 2013 : Computer integrated manufacturing and high speed machining / Abele, Eberhard ; Udiljak, Toma ; Ciglar, Damir - Zagreb : Croatian Association of Production Engineering, 2013, 59-64

ISBN
978-953-7689-02-5

Skup
14th International Scientific Conference on Production Engineering

Mjesto i datum
Biograd, Hrvatska, 19-22.06.2013

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
neural networks; regression; parallel levenberg-marquardt algorithm

Sažetak
In this paper, parallel Levenberg-Marquardt-based feed-forward neural network with variable weight decay, implemented on the Graphics Proce-ssing Unit, is suggested. Two levels of parallelism are implemented in the algorithm. One level of parallelism is achieved across the data set, due to inherently parallel structure of the feed-forward neural networks. Another level of parallelism is achieved in Jacobian computation. To avoid third level of parallelism, i.e. parallelization of optimi-zation search steps, and to keep the algorithm simple, variable decay rate is used. Parameters of variable decay rate rule allow for compromise between oscillations and higher accuracy on one side and stable but slower convergence on the other side. To improve training speed and efficiency modification of random weight initializa-tion is included. Testing of a parallel algorithm is performed on two real domain benchmark problems. Results, given in a form of a table with obtained speedups, show the effectiveness of proposed algorithm implementation.

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo



POVEZANOST RADA


Projekti:
120-1201842-3048 - Umjetna inteligencija u upravljanju složenim nelinearnim dinamičkim sustavima (Kasać, Josip, MZOS ) ( POIROT)
120-1201948-1945 - Inteligentno vođenje obradnih sustava (Majetić, Dubravko, MZOS ) ( POIROT)

Ustanove:
Fakultet strojarstva i brodogradnje, Zagreb

Profili:

Avatar Url Danko Brezak (autor)

Avatar Url Dubravko Majetić (autor)

Avatar Url Tomislav Baček (autor)

Poveznice na cjeloviti tekst rada:

Pristup cjelovitom tekstu rada

Citiraj ovu publikaciju:

Baček, Tomislav; Majetić, Dubravko; Brezak, Danko
Parallel levenberg-marquardt-based neural network with variable decay rate // CIM 2013 : Computer integrated manufacturing and high speed machining / Abele, Eberhard ; Udiljak, Toma ; Ciglar, Damir (ur.).
Zagreb: Croatian Association of Production Engineering, 2013. str. 59-64 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
Baček, T., Majetić, D. & Brezak, D. (2013) Parallel levenberg-marquardt-based neural network with variable decay rate. U: Abele, E., Udiljak, T. & Ciglar, D. (ur.)CIM 2013 : Computer integrated manufacturing and high speed machining.
@article{article, author = {Ba\v{c}ek, Tomislav and Majeti\'{c}, Dubravko and Brezak, Danko}, year = {2013}, pages = {59-64}, keywords = {neural networks, regression, parallel levenberg-marquardt algorithm}, isbn = {978-953-7689-02-5}, title = {Parallel levenberg-marquardt-based neural network with variable decay rate}, keyword = {neural networks, regression, parallel levenberg-marquardt algorithm}, publisher = {Croatian Association of Production Engineering}, publisherplace = {Biograd, Hrvatska} }
@article{article, author = {Ba\v{c}ek, Tomislav and Majeti\'{c}, Dubravko and Brezak, Danko}, year = {2013}, pages = {59-64}, keywords = {neural networks, regression, parallel levenberg-marquardt algorithm}, isbn = {978-953-7689-02-5}, title = {Parallel levenberg-marquardt-based neural network with variable decay rate}, keyword = {neural networks, regression, parallel levenberg-marquardt algorithm}, publisher = {Croatian Association of Production Engineering}, publisherplace = {Biograd, Hrvatska} }




Contrast
Increase Font
Decrease Font
Dyslexic Font