Trevor Gale Matei Zaharia Cliff Young Erich Elsen Sparse GPU kernels for deep learning. 17 2020 SC https://doi.org/10.1109/SC41405.2020.00021 https://dl.acm.org/doi/10.5555/3433701.3433723 conf/sc/2020 db/conf/sc/sc2020.html#GaleZYE20