CuParcone A High-Performance Evolvable Neural Network Model

Author

Chen, Xiaoxi ; Gao, Lin ; De Garis, Hugo

Author_Institution

Xiamen Univ. Sci. & Technol., Xiamen, China

Volume

1

fYear

2010

fDate

11-12 May 2010

Firstpage

1070

Lastpage

1074

Abstract

An algorithm for evolving recurrent neural network via the genetic algorithm was implemented on the CUDA, resulting in a system called CuParcone (CUDA based Partially Connected Neural Evolutionary). Run on a Nvidia Tesla “GPU supercomputer, ” CuParcone achieves a performance increase of 323 times in face gender recognition compared to the comparable Parcone algorithm on a state-of-the-art, commodity single-processor server. The accuracy on this task does not decrease in moving from Parcone to CuParcone, and is comparable to the published results of other algorithms.

Keywords

computer graphic equipment; coprocessors; genetic algorithms; recurrent neural nets; CUDA; CuParcone; GPU supercomputer; Parcone; evolvable neural network model; genetic algorithm; recurrent neural network; Application software; Clustering algorithms; Computer architecture; Computer networks; Concurrent computing; Face recognition; Neural networks; Neurons; Parallel processing; Recurrent neural networks; CUDA; CuParcone; GPU; Gender Recognition; Genetic Algorithms; Neural Networks; Parcone;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Computation Technology and Automation (ICICTA), 2010 International Conference on

Conference_Location

Changsha

Print_ISBN

978-1-4244-7279-6

Electronic_ISBN

978-1-4244-7280-2

Type

conf

DOI

10.1109/ICICTA.2010.479

Filename

5523402