衡量机器学习中的刻板印象危害需要理解谁受到哪些错误以何种方式的伤害
Measuring Machine Learning Harms from Stereotypes Requires Understanding Who Is Harmed by Which Errors in What Ways
AI总结：
本研究通过图像搜索案例，发现强化刻板印象的机器学习错误主要造成主观体验伤害，且对女性影响更大，提出公平性缓解需考虑伤害的具体类型与对象。
摘要：
随着机器学习应用的激增，我们需要理解其潜在的危害。然而，当前的公平性指标很少基于人类对危害的心理体验。借鉴社会心理学中的刻板印象理论，我们以图像搜索中的性别刻板印象为案例研究，考察人们如何对机器学习错误做出反应。首先，我们通过调查研究表明，并非所有机器学习错误都反映刻板印象，也并非同等有害。然后，在实验研究中，我们随机让参与者接触强化刻板印象、违反刻板印象和中性的机器学习错误。我们发现，强化刻板印象的错误会引发更多体验性（即主观性）的伤害，而对认知信念、态度或行为的影响却微乎其微。这种体验性伤害对女性的影响大于男性。然而，某些违反刻板印象的错误对男性更具体验性伤害，这可能是因为对男性气质的感知威胁。我们得出结论，伤害不能作为公平性缓解措施的唯一指导，并提出了一个基于谁在经历何种伤害以及为何经历的细致视角。
英文摘要：
As machine learning applications proliferate, we need an understanding of their potential for harm. However, current fairness metrics are rarely grounded in human psychological experiences of harm. Drawing on the social psychology of stereotypes, we use a case study of gender stereotypes in image search to examine how people react to machine learning errors. First, we use survey studies to show that not all machine learning errors reflect stereotypes nor are equally harmful. Then, in experimental studies we randomly expose participants to stereotype-reinforcing, -violating, and -neutral machine learning errors. We find stereotype-reinforcing errors induce more experientially (i.e., subjectively) harmful experiences, while having minimal changes to cognitive beliefs, attitudes, or behaviors. This experiential harm impacts women more than men. However, certain stereotype-violating errors are more experientially harmful for men, potentially due to perceived threats to masculinity. We conclude that harm cannot be the sole guide in fairness mitigation, and propose a nuanced perspective depending on who is experiencing what harm and why.
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。
