GitHub - JortdeJong13/sam-experiment: In Sharpness-Aware Minimization for Efficiently Improving Generalization the authors propose a new optimization technique, SAM. SAM is not satisfied with just a low loss, instead it seeks a neighbourhood with uniformly low loss. I wanted to try out SAM on the MNIST dataset using a simple CNN

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
SAM_Optimizer.ipynb		SAM_Optimizer.ipynb

About

In Sharpness-Aware Minimization for Efficiently Improving Generalization the authors propose a new optimization technique, SAM. SAM is not satisfied with just a low loss, instead it seeks a neighbourhood with uniformly low loss. I wanted to try out SAM on the MNIST dataset using a simple CNN