Nothing Special   »   [go: up one dir, main page]

CERN Accelerating science

CERN Document Server Намерени са 1 записа  Търсенето отне 0.61 секунди. 
1.
Evaluating Mixed-Precision Arithmetic for 3D Generative Adversarial Networks to Simulate High Energy Physics Detectors / Ríos, John Osorio (Barcelona, Polytechnic U.) ; Armejach, Adrià (Barcelona, Polytechnic U.) ; Khattak, Gulrukh (CERN) ; Petit, Eric (Unlisted, FR) ; Vallecorsa, Sofia (CERN) ; Casas, Marc (Barcelona, Polytechnic U.)
Several hardware companies are proposing native Brain Float 16-bit (BF16) support for neural network training. The usage of Mixed Precision (MP) arithmetic with floating-point 32-bit (FP32) and 16-bit half-precision aims at improving memory and floating-point operations throughput, allowing faster training of bigger models. [...]
2020 - 8 p. - Published in : 10.1109/ICMLA51294.2020.00017
In : 2020 19th IEEE International Conference on Machine Learning and Applications, Online, 14 - 17 Dec 2020, pp.49-56

Виж също: автори с подобни имена
43 Casas, M
Интересувате ли се от уведомяване за нови резултати по тази заявка?
Set up a personal email alert or subscribe to the RSS feed.
Не намерихте, каквото търсите? Опитайте търсене в други сървъри:
Casas, Marc в Amazon
Casas, Marc в CERN EDMS
Casas, Marc в CERN Intranet
Casas, Marc в CiteSeer
Casas, Marc в Google Books
Casas, Marc в Google Scholar
Casas, Marc в Google Web
Casas, Marc в IEC
Casas, Marc в IHS
Casas, Marc в INSPIRE
Casas, Marc в ISO
Casas, Marc в KISS Books/Journals
Casas, Marc в KISS Preprints
Casas, Marc в NEBIS
Casas, Marc в SLAC Library Catalog
Casas, Marc в Scirus