A Stochastic Rounding-Enabled Low-Precision Floating-Point MAC for DNN Training.

Sami Ben Ali Silviu-Ioan Filip Olivier Sentieys

Published in: CoRR (2024)

Keyphrases

floating point
training process
fixed point
square root
instruction set
sparse matrices
training data
linear programming
fast fourier transform
floating point arithmetic
fine grained
approximation algorithms
training set
pairwise
probabilistic model
markov random field