Jul 2026
Leveraging Extragradient for Effective Sharpness-Aware Minimization in Deep Learning
Theoretical analysis further confirms that EISAM tightens the generalization bound by steering parameters toward flatter minima with reduced curvature, establishing it as a robust, scalable, and broadly applicable optimization solution that advances both the theory and practice in deep learning.
Yao Fu, Chunxia Zhang, Junmin Liu et al.
· arXiv.org · 0 citations