Automated Learning Rate Scheduler for Large-batch Training

Automated Learning Rate Scheduler for Large-batch Training

https://arxiv.org/abs/2107.05855

Sep 16, 2021

Anomaly Detection,

ICML (2021)

1. どんなもの?

2. 先行研究と比べてどこがすごい?

3. 技術や手法の”キモ”はどこ?

変数定義

ウォームアップ段階

減衰段階

4. どうやって有効だと検証した?

5. 議論はあるか?

6. 関連文献

  1. Byeongho Heo, Sanghyuk Chun, Seong Joon Oh, Dongyoon Han, Sangdoo Yun, Gyuwan Kim, Youngjung Uh, and Jung-Woo Ha. Adamp: Slowing down the slowdown for momentum optimizers on scale-invariant weights. In International Conference on Learning Representations, 2021. URL https://openreview.net/forum?id=Iz3zU3M316D.
一覧へ戻る