CREW: Computation reuse and efficient weight storage for hardware-accelerated MLPs and RNNs.
Marc RieraJosé-María ArnauAntonio GonzálezPublished in: J. Syst. Archit. (2022)
Keyphrases
- storage cost
- artificial neural networks
- efficient computation
- data retrieval
- cost effective
- software reuse
- recurrent neural networks
- computationally efficient
- databases
- computationally expensive
- lightweight
- decision support system
- multi layer perceptron
- learning objects
- limited memory
- information systems
- data mining