@article{acceleratinglargebatchtrainingvia,
  title = {Accelerating Large Batch Training via Gradient Signal to Noise Ratio (GSNR)},
  author = {Guo-qing Jiang and Jinlong Liu and Zixiang Ding and Lin Guo and Wei Lin},
  year = {2023},
  eprint = {2309.13681},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2309.13681v1},
}