Compact CNN Training Accelerator with Variable Floating-Point Datapath.

Jiun Hong TaeGeon Lee Saad Arslan HyungWon Kim

Published in: ISOCC (2020)

Keyphrases

floating point
fixed point
square root
cellular neural networks
sparse matrices
instruction set
floating point arithmetic
training set
fine grained
interval arithmetic
neural network
similarity measure
higher order
parallel implementation
convolutional neural network