Optimizing Weight Mapping and Data Flow for Convolutional Neural Networks on Processing-in-Memory Architectures.
Xiaochen PengRui LiuShimeng YuPublished in: IEEE Trans. Circuits Syst. I Fundam. Theory Appl. (2020)
Keyphrases
- data flow
- convolutional neural networks
- data stream processing
- data transfer
- database machine
- control flow
- memory management
- general purpose processors
- digital signal processors
- computational power
- object oriented software
- memory requirements
- information processing
- memory hierarchy
- systolic array
- real time
- object oriented
- random access
- digital signal processing
- data processing
- feature extraction
- case study
- parallel processing
- main memory
- parallel algorithm
- bottle neck