Using GPUs for machine learning algorithms
D. Steinkraus, Ian Buck, P.Y. Simard · Proceedings of the International Conference on Document Analysis and Recognition · 2005
Using dedicated hardware to do machine learning typically ends up in disaster because of cost, obsolescence, and poor software. The popularization of graphic processing units (GPUs), which are now available on every PC, provides an attractive alternative. We propose a generic 2-layer fully connected neural network GPU implementation which yields over 3/spl times/ speedup for both training and testing with respect to a 3 GHz P4 CPU.
Cited by 238 works. doi:10.1109/icdar.2005.251
Authors
References
- Best practices for convolutional neural networks applied to visual document analysis
- Sparse matrix solvers on the GPU
- Sparse matrix solvers on the GPU
- Linear algebra operators for GPU implementation of numerical algorithms
- Ray tracing on programmable graphics hardware
- Ray tracing on programmable graphics hardware
- The GPU enters computing's mainstream
- Linear algebra operators for GPU implementation of numerical algorithms