Task Parallelism-Aware Deep Neural Network Scheduling on Multiple Hybrid Memory Cube-Based Processing-in-Memory.
Young Sik LeeTae Hee HanPublished in: IEEE Access (2021)
Keyphrases
- neural network
- computational power
- auto associative
- processing elements
- memory usage
- artificial neural networks
- scheduling problem
- random access
- memory requirements
- memory management
- back propagation
- associative memory
- memory space
- parallel processors
- computing power
- main memory
- memory size
- pattern recognition
- compute intensive
- parallel computation
- feed forward
- data processing
- data management
- multi dimensional
- information systems