BriefGPT.xyz
Dec, 2021
隐含层次和多组学习的简单近似最优算法
Simple and near-optimal algorithms for hidden stratification and multi-group learning
HTML
PDF
Christopher Tosh, Daniel Hsu
TL;DR
本论文研究了多组学习问题的解决结构并提供了简单和近乎最优的算法，以解决子群体的条件风险与公平性问题。
Abstract
multi-group agnostic learning
is a formal learning criterion that is concerned with the
conditional risks
of predictors within
subgroups
o
→