CuParcone A High-Performance Evolvable Neural Network Model
Xiamen Univ. Sci. & Technol., Xiamen, China
International Conference on Intelligent Computation Technology and Automation (ICICTA), 2010
@inproceedings{chen2010cuparcone,
title={CuParcone A High-Performance Evolvable Neural Network Model},
author={Chen, X. and Gao, L. and de Garis, H.},
booktitle={Proceedings of the 2010 International Conference on Intelligent Computation Technology and Automation-Volume 01},
pages={1070–1074},
year={2010},
organization={IEEE Computer Society}
}
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.
June 14, 2011 by hgpu
