Sharpness-Aware Minimization

2000년 도입 · 논문 142편에서 사용

Sharpness-Aware Minimization, or SAM, is a procedure that improves model generalization by simultaneously minimizing loss value and loss sharpness. SAM functions by seeking parameters that lie in neighborhoods having uniformly low loss value (rather than parameters that only themselves have low loss value).

출처: Sharpness-Aware Minimization for Efficiently Improving Generalization

소개 논문: Sharpness-Aware Minimization for Efficiently Improving Generalization