Abstract | ||
---|---|---|
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 3X speedup for both training and testing with respect to a 3GHz P4 CPU. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1109/ICDAR.2005.251 | ICDAR-1 |
Keywords | Field | DocType |
machine learning algorithms,attractive alternative,neural network gpu implementation,dedicated hardware,poor software,p4 cpu,graphic processing units,machine learning,neural network,learning artificial intelligence,neural nets | Obsolescence,Active learning (machine learning),Computer science,Software,Artificial intelligence,Computational learning theory,Artificial neural network,Cellular neural network,Machine learning,Speedup | Conference |
ISSN | ISBN | Citations |
1520-5363 | 0-7695-2420-6 | 48 |
PageRank | References | Authors |
5.40 | 5 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dave Steinkrau | 1 | 48 | 5.40 |
Patrice Y. Simard | 2 | 1112 | 155.00 |
Ian Buck | 3 | 1544 | 149.20 |