Breaking MLPerf Training: A Case Study on Optimizing BERT.
Yongdeok KimJaehyung AhnMyeongwoo KimChangin ChoiHeejae KimNarankhuu TuvshinjargalSeungwon LeeYanzi ZhangYuan PeiXiongzhan LinghuJingkun MaLin ChenYuehua DaiSungjoo YooPublished in: CoRR (2024)