Parallel Batch Training of the Self-Organizing Map Using OpenCL
ViSLAB, School of Information Technologies, The University of Sydney, NSW, Australia
Neural Information Processing. Models and Applications, Lecture Notes in Computer Science, 2010, Volume 6444/2010, 470-476
@article{takatsuka2010parallel,
title={Parallel batch training of the self-organizing map using openCL},
author={Takatsuka, M. and Bui, M.},
journal={Neural Information Processing. Models and Applications},
pages={470–476},
year={2010},
publisher={Springer}
}
The Self-Organizing Maps (SOMs) are popular artificial neural networks that are often used for data analyses through clustering and visualisation. SOM’s mathematical model is inherently parallel. However, many implementations have not successfully exploited its parallelism because previous attempts often required cluster-like infrastructures. This article presents the parallel implementation of SOMs, particularly the batch map variant using Graphics Processing Units (GPUs) through the use of Open Computing Language (OpenCL).
August 18, 2011 by hgpu
