FPGA-based training of convolutional neural networks with a reduced precision floating-point library.

Roberto DiCecco Lin Sun Paul Chow

Published in: FPT (2017)

Keyphrases

floating point
convolutional neural networks
fixed point
square root
fast fourier transform
convolutional network
instruction set
data processing
sparse matrices
floating point arithmetic
computer vision
training set
general purpose
embedded systems
application specific
interval arithmetic