AN ADAPTIVE POLICY TO EMPLOY SHARPNESS-AWARE MINIMIZATION

作     者：Jiang, Weisen Yang, Hansi Zhang, Yu Kwok, James 

作者机构：Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong Peng Cheng Laboratory China 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2023年

核心收录：

摘      要：Sharpness-aware minimization (SAM), which searches for flat minima by min-max optimization, has been shown to be useful in improving model generalization. However, since each SAM update requires computing two gradients, its computational cost and training time are both doubled compared to standard empirical risk minimization (ERM). Recent state-of-the-arts reduce the fraction of SAM updates and thus accelerate SAM by switching between SAM and ERM updates randomly or periodically. In this paper, we design an adaptive policy to employ SAM based on the loss landscape geometry. Two efficient algorithms, AE-SAM and AE-LookSAM, are proposed. We theoretically show that AE-SAM has the same convergence rate as SAM. Experimental results on various datasets and architectures demonstrate the efficiency and effectiveness of the adaptive policy. © 2023, CC BY.