Reviewer 1 of these papers investigate the relationship between regret and stability of an online learning algorithm and a comparison