Pattern recognition with OpenCL heterogeneous platform
作者： Jordan Vrtanoski , Toni Draganov Stojanovski
DOI: 10.1109/TELFOR.2012.6419306
关键词:
摘要: OpenCL platform provides unified development environment for various multicore processors. In this paper, we evaluate the framework application in pattern recognition. We have selected most common algorithm Artificial Neural Networks (ANN) training — backpropagation parallelization with because of its high demand processing resources. will show a SIMD version suitable implementation. Our implementation showed 25.8 speedup execution on ATI 5870 GPU compared to Intel Xeon W3530 when MNIST handwritten digits data set.
sci-hub.se 参考文章(10)
D. Jakobovic, N. Krpan, Parallel neural network training with OpenCL international convention on information and communication technology, electronics and microelectronics. pp. 1053- 1057 ,(2012)
Neil G. Dickson, Kamran Karimi, Firas Hamze, Importance of explicit vectorization for CPU and GPU software performance Journal of Computational Physics. ,vol. 230, pp. 5383- 5398 ,(2011) , 10.1016/J.JCP.2011.03.041
Li Deng, The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web] IEEE Signal Processing Magazine. ,vol. 29, pp. 141- 142 ,(2012) , 10.1109/MSP.2012.2211477
F. Rosenblatt, The perceptron: a probabilistic model for information storage and organization in the brain. Psychological Review. ,vol. 65, pp. 386- 408 ,(1958) , 10.1037/H0042519
K. Fatahalian, J. Sugerman, P. Hanrahan, Understanding the efficiency of GPU algorithms for matrix-matrix multiplication siggraph eurographics conference on graphics hardware. pp. 133- 137 ,(2004) , 10.1145/1058129.1058148
Byunghyun Jang, Dana Schaa, Perhaad Mistry, David Kaeli, Exploiting Memory Access Patterns to Improve Memory Performance in Data-Parallel Architectures IEEE Transactions on Parallel and Distributed Systems. ,vol. 22, pp. 105- 118 ,(2011) , 10.1109/TPDS.2010.107
Mark J. Harris, William V. Baxter, Thorsten Scheuermann, Anselmo Lastra, Simulation of cloud dynamics on graphics hardware ACM SIGGRAPH 2005 Courses on - SIGGRAPH '05. pp. 223- ,(2005) , 10.1145/1198555.1198793
Victor G. Tsaregorodtsev, Parallel implementation of back-propagation neural network software on SMP computers parallel computing technologies. pp. 186- 192 ,(2005) , 10.1007/11535294_16
Xavier Sierra-Canto, Francisco Madera-Ramirez, Victor Uc-Cetina, Parallel Training of a Back-Propagation Neural Network Using CUDA international conference on machine learning and applications. pp. 307- 312 ,(2010) , 10.1109/ICMLA.2010.52
Yu-Chi Ho, Arthur E. Bryson, Applied optimal control : optimization, estimation and control Applied optimal control : optimization, estimation, and control / Arthur E. Bryson, Jr., Yu-Chi Ho.. ,(1975)