Robust and Heavy-Tailed Mean Estimation Made Simple, via Regret Minimization
NIPS 2020, 2020.
We showed that in robust and heavy-tailed settings, the problem can be approached by techniques from regret minimization and online learning
We study the problem of estimating the mean of a distribution in high dimensions when either the samples are adversarially corrupted or the distribution is heavy-tailed. Recent developments in robust statistics have established efficient and (near) optimal procedures for both settings. However, the algorithms developed on each side tend...More
PPT (Upload PPT)