Computing Bias · 计算偏见
| English | 中文 | Pinyin · 拼音 |
|---|---|---|
| Computing bias/kəmˈpjuːtɪŋ ˈbaɪəs/ | 计算偏见 | jì suàn piān jiàn |
| training data/ˈtreɪnɪŋ ˈdeɪtə/ | 训练数据 | xùn liàn shù jù |
| diverse/daɪˈvɜːs/ | 多样的 | duō yàng de |
Unfairness built into a program
- Computing bias 计算偏见 is unfairness built into a program — sometimes without developers even realising it.
- A program can treat some groups worse than others.
- This is not always deliberate; it can be an accident of how the program was built.
- Bias matters because software increasingly makes real decisions about people.
内建在程序里的不公
- 计算偏见(computing bias)是内建在程序里的不公——有时开发者甚至没意识到。
- 一个程序能比对待其他群体更差地对待一些群体。
- 这不总是故意的;它可以是程序如何被构建的一个意外。
- 偏见重要,因为软件越来越多地对人做出真实的决定。
Computing bias is: · 计算偏见是:
It can treat some groups worse, even by accident. · 它能更差地对待一些群体,甚至是意外。
Bias comes from the data
- A major source is the data used to build the program.
- Many modern programs learn from training data 训练数据 — examples they study to find patterns.
- If that data is biased, the program's results are biased too.
- Biased training data leads to unfair results for the under-represented group.
偏见来自数据
- 一个主要来源是用来构建程序的数据。
- 许多现代程序从训练数据(training data)学习——它们研究以找到模式的例子。
- 如果那些数据有偏差,程序的结果也有偏差。
- 有偏差的训练数据导致对代表不足的群体的不公结果。
How bias enters a program · 偏见如何进入一个程序
Biased training data leads a program to learn skewed patterns and produce unfair results; diverse testing reveals it, and representative data reduces it. · 有偏差的训练数据让程序学到倾斜的模式并产生不公的结果;多样测试揭示它,有代表性的数据减少它。
Programs that learn from examples study their ______ data to find patterns. · 从例子学习的程序研究它们的______数据来找模式。
Biased training data gives biased results. · 有偏差的训练数据给出有偏差的结果。
A major source of computing bias is: · 计算偏见的一个主要来源是:
If the data is skewed, the program's results are skewed. · 如果数据是倾斜的,程序的结果是倾斜的。
Testing with diverse users
- Testing with diverse 多样的 users and data helps reveal hidden bias before release.
- If everyone who tests looks the same, the bias stays invisible.
- A varied test set exposes where the program fails different groups.
- This is why diverse testing is a core fix, not an optional extra.
用多样的用户测试
- 用多样的(diverse)用户和数据测试,有助于在发布之前揭示隐藏的偏见。
- 如果测试的每个人看起来都一样,偏见就保持不可见。
- 一个多样的测试集暴露程序在哪里让不同群体失望。
- 这就是为什么多样测试是核心的修复,而非可选的额外项。
Testing with a diverse set of users helps reveal hidden bias before release. · 用一个多样的用户集测试有助于在发布前揭示隐藏的偏见。
A varied test set exposes failures on different groups. · 一个多样的测试集暴露对不同群体的失败。
A face system trained mostly on light-skinned men makes more errors on others because: · 一个主要用浅肤色男性训练的人脸系统对其他人犯更多错,因为:
The under-represented groups get worse results. · 代表不足的群体得到更差的结果。
Developers can reduce bias by: · 开发者能通过什么减少偏见:
Representative data and cross-group checks make results fairer. · 有代表性的数据和跨群体检查让结果更公平。
Reducing bias
- Developers can reduce bias by using representative data and checking results across different groups.
- The goal is a program that works fairly for everyone, not just the majority.
Face recognition. A system trained mostly on light-skinned men works well on them but makes far more mistakes on women and darker-skinned people — it saw fewer such faces while learning. The developers did not intend this; the bias came from unrepresentative training data. Testing with a diverse set of faces would have exposed it.
减少偏见
- 开发者能通过使用有代表性的数据并跨不同群体检查结果来减少偏见。
- 目标是一个对每个人都公平工作的程序,而非只对多数。
人脸识别。 一个主要用浅肤色男性训练的系统对他们工作良好,但对女性和深肤色的人犯错多得多——它在学习时看到的这类脸更少。开发者并非有意;偏见来自不具代表性的训练数据。用一个多样的脸集测试本会暴露它。
Computing bias is unfairness built into a program, often unintentionally. A major source is biased training data — if the examples are skewed, so are the results. Testing with diverse users reveals hidden bias before release, and using representative data and checking across groups reduces it.
计算偏见是内建在程序里的不公,常常无意。一个主要来源是有偏差的训练数据——如果例子是倾斜的,结果也是。用多样的用户测试在发布前揭示隐藏的偏见,使用有代表性的数据并跨群体检查减少它。