MALOC: A Fully Pipelined FPGA Accelerator for Convolutional Neural Networks With All Layers Mapped on Chip.
Lei GongChao WangXi LiHuaping ChenXuehai ZhouPublished in: IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. (2018)
Keyphrases
- convolutional neural networks
- field programmable gate array
- programmable logic
- high speed
- single chip
- low cost
- low power consumption
- hardware implementation
- convolutional network
- parallel architecture
- application specific integrated circuits
- embedded systems
- digital signal processors
- reconfigurable hardware
- low power
- parallel implementation
- systolic array
- hardware architecture
- data flow
- parallel computing
- fpga device
- image processing algorithms
- fpga implementation
- digital signal processing
- computing systems
- real time image processing
- real time
- vlsi implementation
- physical design
- analog vlsi
- hardware design
- computer systems
- massively parallel
- high density