注册并分享邀请链接,可获得视频播放与邀请奖励。

Fengzhuo Zhang 的个人资料封面
Fengzhuo Zhang 的头像

Fengzhuo Zhang (@FengzhuoZhang)

@FengzhuoZhang
Postdoc @Yale @YaleCADMY |Previous: ECE Ph.D. @NUSingapore | EE undergrad @Tsinghua_Uni
424 正在关注    403 粉丝
Beyond faster training, does Muon learn better features than Adam? 🚀 Ans: Yes. Muon learns features that are more robust to input corruptions and transfer better to downstream tasks. This advantage is reflected in hidden states: 1⃣larger logit margins → stronger robustness 2⃣higher effective rank → richer, more transferable representations Paper Link: A thread 🧵
显示更多