arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.05902cs.LGcs.AIcs.CLcs.CY
收藏

谁的偏好?公平偏好的差异及其对利用人类反馈的人工智能公平性的影响

Whose Preferences? Differences in Fairness Preferences and Their Impact on the Fairness of AI Utilizing Human Feedback

ETH Zürich(苏黎世联邦理工学院);MPI for Intelligent Systems(马克斯·普朗克智能系统研究所);University of Oxford(牛津大学)

Emilia Agis Lerner, Florian E. Dorner, Elliott Ash, Naman Goel

AI总结:

本文通过新数据集揭示不同人口统计学群体在内容审核公平偏好上的显著差异,并提出对不同群体分类器等权集成的方案以改善不同人口统计学交叉群体的公平性表现。

摘要:

关于从人类反馈中学习以使机器学习系统的各个方面与人类价值观和偏好保持一致的研究正在不断增加。我们考虑内容审核中的公平性设置,其中利用人类反馈来确定两条评论——分别涉及不同的敏感属性群体——应如何相互比较处理。通过从Prolific和MTurk收集的新数据集,我们发现公平偏好因标注者的种族、年龄、政治立场、教育水平和LGBTQ+身份而存在显著差异。我们还证明,文本中提及的人口统计学特征对用户如何感知审核中的个体公平性有强烈影响。此外,我们发现,在训练用于预测人类偏好的下游分类器中也存在差异。最后,我们观察到,与对每条标注赋予相等权重的单一分类器相比,对基于不同人口统计学群体的标注训练的分类器赋予相等权重的集成方法,在不同人口统计学交叉群体上表现更好。

英文摘要:

There is a growing body of work on learning from human feedback to align various aspects of machine learning systems with human values and preferences. We consider the setting of fairness in content moderation, in which human feedback is used to determine how two comments -- referencing different sensitive attribute groups -- should be treated in comparison to one another. With a novel dataset collected from Prolific and MTurk, we find significant gaps in fairness preferences depending on the race, age, political stance, educational level, and LGBTQ+ identity of annotators. We also demonstrate that demographics mentioned in text have a strong influence on how users perceive individual fairness in moderation. Further, we find that differences also exist in downstream classifiers trained to predict human preferences. Finally, we observe that an ensemble, giving equal weight to classifiers trained on annotations from different demographics, performs better for different demographic intersections; compared to a single classifier that gives equal weight to each annotation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑