A 3.89-GOPS/mW Scalable Recurrent Neural Network Processor With Improved Efficiency on Memory and Computation.
Jiaquan WuFeiteng LiZhijian ChenXiaoyan XiangPublished in: IEEE Trans. Very Large Scale Integr. Syst. (2019)
Keyphrases
- recurrent neural networks
- memory usage
- complex valued
- neural network
- feed forward
- echo state networks
- recurrent networks
- power consumption
- memory management
- limited memory
- reservoir computing
- feedforward neural networks
- neural model
- artificial neural networks
- hidden layer
- back propagation
- highly efficient
- multithreading
- memory hierarchy
- operating system
- computation intensive
- intel xeon
- fuzzy neural network
- high scalability
- high speed