ALAM: Averaged Low-Precision Activation for Memory-Efficient Training of Transformer Models.

Sunghyeon Woo Sunwoo Lee Dongsuk Jeon

Published in: ICLR (2024)

Keyphrases

memory efficient
neural network
artificial intelligence
search algorithm
fuzzy logic
information processing
power system
prior knowledge
probabilistic model
model selection
parameter estimation
complex systems
statistical model
neural network model
historical data