Implementation of Fully-Pipelined CNN Inference Accelerator on FPGA and HBM2 Platform.
Van-Cam NguyenYasuhiko NakashimaPublished in: IEICE Trans. Inf. Syst. (2023)
Keyphrases
- parallel architecture
- field programmable gate array
- reconfigurable hardware
- hardware implementation
- real time
- hardware architecture
- low cost
- xilinx virtex
- software implementation
- high speed
- efficient implementation
- fpga implementation
- parallel implementation
- neural network
- dedicated hardware
- hardware design
- inference process
- parallel processing
- multi agent systems
- single chip
- platform independent
- computing platform
- probabilistic inference
- layered architecture
- bayesian networks
- pilot testing