Model Generalization: A Sharpness Aware Optimization Perspective

TL;DR

This paper investigates how sharpness-aware optimization methods like SAM and ASAM enhance model generalization, demonstrating their effectiveness through experiments and highlighting areas for further research.

Contribution

The paper proposes three experiments to validate the generalization benefits of sharpness-aware optimization techniques, especially focusing on ASAM's performance on un-normalized data.

Findings

Sharpness-aware methods improve model generalization.

ASAM enhances performance on un-normalized data.

Further research needed for ASAM's robustness.

Abstract

Sharpness-Aware Minimization (SAM) and adaptive sharpness-aware minimization (ASAM) aim to improve the model generalization. And in this project, we proposed three experiments to valid their generalization from the sharpness aware perspective. And our experiments show that sharpness aware-based optimization techniques could help to provide models with strong generalization ability. Our experiments also show that ASAM could improve the generalization performance on un-normalized data, but further research is needed to confirm this.