Zobrazeno 1 - 1
of 1
pro vyhledávání: '"Revanth, Challapalli Phanindra"'
Computing the loss gradient via backpropagation consumes considerable energy during deep learning (DL) model training. In this paper, we propose a novel approach to efficiently compute DL models' gradients to mitigate the substantial energy overhead
Externí odkaz:
http://arxiv.org/abs/2406.07332
