人工智能系统中的偏见
Bias in AI Systems
如果训练数据带有偏见,人工智能也会带有偏见。比如,人脸识别对某些肤色的人效果更好,翻译会把某些词默认为特定性别。我们还会了解偏见从何而来,以及能不能解决它。
If training data is biased, AI will be biased; examples: facial recognition working better for some skin tones, translation assuming gender; where bias comes from and whether we can fix it
学习证据 / Evidence
- 用一个现实中的例子解释人工智能中的偏见是什么意思。
- 说明带有偏见的训练数据会怎样导致人工智能产生有偏见的结果。
- 提出一种减少人工智能系统偏见的方法(比如使用更多样化的数据,或者用不同人群来测试)。
- Explain what bias in AI means using a real-world example
- Describe how biased training data leads to biased AI results
- Suggest one way to reduce bias in an AI system (use more diverse data, test with different groups)
评估问题 / Assessment
{{name}}能解释一下,为什么主要用浅肤色人脸照片训练出来的人工智能,对深肤色的人可能效果不太好?
Could {{name}} explain why an AI trained mostly on photos of light-skinned faces might not work as well for people with darker skin?