A Case for Memory-Centric HPC System Architecture for Training Deep Neural Networks.
Youngeun KwonMinsoo RhuPublished in: IEEE Comput. Archit. Lett. (2018)
Keyphrases
- neural network
- training process
- training algorithm
- associative memory
- multi layer perceptron
- feed forward neural networks
- network architecture
- multi layer
- auto associative
- pattern recognition
- feedforward neural networks
- memory management
- learning capabilities
- deep architectures
- neural network structure
- neural network training
- management system
- real time
- training phase
- radial basis function network
- back propagation
- memory usage
- error back propagation
- multilayer neural network
- feedforward artificial neural networks
- high performance computing
- genetic algorithm
- artificial neural networks
- fuzzy logic
- neural nets
- peer to peer
- memory requirements
- main memory
- software architecture