知识卡片

普通读书笔记卡 · 6527.a · 生成模型:GPT-6

去掉敏感字段仍可能保留代理

理解等级:中级

删除性别等显式输入不一定阻止模型利用姓名等相关代理复制历史差异。需要检查数据生成、标签和可用变量如何携带信息;本书情景用于说明机制,不构成针对真实个人或机构的风险判断和处置建议。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      remove gender labels from the training data. But even if those sexist loan officers didn’t see an explicit indication of the person’s gender, they could infer it from elements of the application, such as the person’s name. Those loan officers encode their prejudicial views in the form of a statistical correlation between those proxy variables for gender and loan outcome. The machine learning algorithm would

    参考来源

    - 位置:《Causal AI》第01章《为何需要因果AI》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-1.xhtml) - 结论依据:摘录中关于去掉敏感字段仍可能保留代理的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:remove gender labels from the training data. But even if those sexist loan officers didn’t see an explicit indication of the person’s gender, they could infer it from elements of the application, such as the person’s name. Those loan officers encode their prejudicial views in the form of a statistical correlation between those proxy variables for gender and loan outcome. The machine learning algorithm would