Training Quantised Neural Networks with STE Variants: the Additive Noise Annealing Algorithm.

Matteo Spallanzani Gian Paolo Leonardi Luca Benini

Published in: CVPR (2022)

Keyphrases

additive noise
neural network
annealing algorithm
noisy images
maximum likelihood
spatial domain
random variables
pattern recognition
simulated annealing
protein folding
deterministic annealing
training set
computer vision
objective function
probability distribution
denoising