Using GPI-2 for Distributed Memory Paralleliziation of the Caffe Toolbox to Speed up Deep Neural Network Training.
Martin KuehnJanis KeuperFranz-Josef PfreundtPublished in: CoRR (2017)
Keyphrases
- distributed memory
- neural network training
- training algorithm
- neural network
- shared memory
- ibm sp
- parallel implementation
- optimization method
- matrix multiplication
- parallel machines
- parallel architecture
- back propagation
- parallel computers
- parallel algorithm
- particle swarm optimisation
- message passing
- support vector machine
- dynamic programming
- artificial neural networks