BriefGPT.xyz
Sep, 2021
公正感知的类别不平衡学习
Fairness-aware Class Imbalanced Learning
HTML
PDF
Shivashankar Subramanian, Afshin Rahimi, Timothy Baldwin, Trevor Cohn, Lea Frermann
TL;DR
探索使用长尾学习方法对推特情感和职业分类中的类别不平衡和偏见进行干预，并扩展了基于边际损失的方法以实现公平性。经过受控实验，实证表明所提出的方法有助于缓解类别不平衡和人口统计学偏见。
Abstract
class imbalance
is a common challenge in many
nlp
tasks, and has clear connections to
bias
, in that
→