, 3.08Gop/s/mW @ 1.2 mW accelerator for near-sensor recurrent neural network inference.
Francesco ContiLukas CavigelliGianna PaulinIgor SusmeljLuca BeniniPublished in: CICC (2018)
Keyphrases
- recurrent neural networks
- power consumption
- power supply
- complex valued
- neural network
- recurrent networks
- feed forward
- power plant
- long short term memory
- neural model
- real time
- feedforward neural networks
- artificial neural networks
- echo state networks
- hidden layer
- reservoir computing
- inter frame
- image quality
- sensor networks
- video streaming
- machine learning