Automatic Cross-Replica Sharding of Weight Update in Data-Parallel TrainingXu Yuanzhong,Lee HyoukJoong,Chen Dehao,Choi Hongjun,Hechtman Blake,Wang ShiboarXiv (Cornell University)(2020)引用 15|浏览127关键词weight update,training,cross-replica,data-parallelAI 理解论文溯源树样例生成溯源树,研究论文发展脉络Chat Paper正在生成论文摘要