TY - RPRT TI - Towards Low-Overhead Resilience for Data Parallel Deep Learning AU - Nicolae, Bogdan AU - Hobson, Tanner AU - Yildiz, Orcun AU - Peterka, Thomas AU - Morozov, Dmitriy PY - 2022 DO - 10.1109/ccgrid54584.2022.00043 UR - https://www.osti.gov/biblio/1887187 ID - 1887187 ER -