A Scalable Near-Memory Architecture for Training Deep Neural Networks on Large In-Memory Datasets.
Fabian SchuikiMichael SchaffnerFrank K. GürkaynakLuca BeniniPublished in: CoRR (2018)
Keyphrases
- neural network
- associative memory
- memory usage
- training process
- limited memory
- memory requirements
- memory size
- real time
- pattern recognition
- low memory
- auto associative
- network architecture
- computing power
- memory space
- memory subsystem
- memory hierarchy
- memory management
- radial basis function network
- random access
- training dataset
- genetic algorithm
- main memory
- benchmark datasets
- supervised learning