Ultra-low latency recurrent neural network inference on FPGAs for physics applications with hls4ml.
Elham E KhodaDylan S. RankinRafael Teixeira de LimaPhilip C. HarrisScott HauckShih-Chieh HsuMichael KaganVladimir LoncarChaitanya PaikaraRicha RaoSioni SummersCaterina VernieriAaron WangPublished in: Mach. Learn. Sci. Technol. (2023)
Keyphrases
- recurrent neural networks
- low latency
- high speed
- high bandwidth
- feed forward
- neural network
- high throughput
- recurrent networks
- complex valued
- reservoir computing
- virtual machine
- echo state networks
- massive scale
- real time
- highly efficient
- stream processing
- artificial neural networks
- continuous query processing
- field programmable gate array
- data acquisition
- artificial intelligence
- data sets