NNPIM: A Processing In-Memory Architecture for Neural Network Acceleration.
Saransh GuptaMohsen ImaniHarveen KaurTajana Simunic RosingPublished in: IEEE Trans. Computers (2019)
Keyphrases
- neural network
- associative memory
- processing elements
- real time
- memory management
- auto associative
- network architecture
- neural architecture
- random access
- parallel architecture
- data processing
- distributed processing
- computational power
- genetic algorithm
- multi layer
- hardware architecture
- artificial neural networks
- computing power
- hardware implementation
- fault diagnosis
- parallel computers
- neural network is trained
- memory size
- expert systems
- learning capabilities
- reconfigurable hardware
- direct memory access
- limited memory
- design considerations
- operating system
- memory usage
- fuzzy logic
- recurrent neural networks
- multilayer perceptron
- feed forward
- neural network model
- main memory
- software architecture
- back propagation
- information processing