DocumentCode :
2226102
Title :
Self-Organizing Maps Neural Networks on Parallel Cluster
Author :
Zhu Liping ; Guo Wensheng ; Bai Yanping
Author_Institution :
Dept. of Comput. Sci. & Technol., China Univ. of Pet.-Beijing, Beijing, China
fYear :
2009
fDate :
26-28 Dec. 2009
Firstpage :
384
Lastpage :
388
Abstract :
Parallel design and realization of artificial neutral networks on clusters can fully employ the advantage of ANN parallel processing, shorten the training time and reduce the algorithm complexity. As the parallel techniques become more and more developed, it´s increasingly important to design artificial neutral networks on clusters via a combined software and hardware method. The support of parallel cluster technique on artificial neutral networks is discussed in several aspects on software and hardware platforms. A design method and basis frame of SOM neutral networks on parallel clusters is presented, and problems on the training efficiency of neutral networks on parallel clusters are further discussed. This approach is widely applicable to the computer realization on many neutral network models.
Keywords :
artificial intelligence; parallel processing; self-organising feature maps; workstation clusters; ANN parallel processing; SOM neutral networks; algorithm complexity; artificial neutral networks; parallel cluster technique; self-organizing map neural networks; Computational modeling; Computer networks; Computer science; Concurrent computing; Design methodology; Ethernet networks; Hardware; Neural networks; Parallel processing; Self organizing feature maps;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Science and Engineering (ICISE), 2009 1st International Conference on
Conference_Location :
Nanjing
Print_ISBN :
978-1-4244-4909-5
Type :
conf
DOI :
10.1109/ICISE.2009.1053
Filename :
5455264
Link To Document :
بازگشت