← Search

Dongkuk Si

2 accepted papers

2026

Minor First, Major Last: A Depth-Induced Implicit Bias of Sharpness-Aware Minimization

ICLR 2026poster

We study the implicit bias of sharpness-aware minimization (SAM) when training $L$-layer linear diagonal networks on linearly separable binary classification. For linear models ($L=1$), both $\ell_\infty$- and $\ell_2$-SAM recover the $\ell_2$ max-margin classifier, matching gradient descent (GD). H…

Cited by 0SourceScholar