[3GPM2] 機械学習(3)

4:30 PM - 4:50 PM JST(7:30 AM - 7:50 AM UTC)

[3GPM2-04] Transformers Learn Nonlinear Features In Context: Nonconvex Mean-field Dynamics on the Attention Landscape

〇キム ジュノ^1,2、鈴木 大慈^1,2 (1. 東京大学、2. 理化学研究所)

Password required to view