Bias and Fairness

Readings

• (Skim) Solon Barocas, Moritz Hardt, Arvind Narayanan, “Fairness and Machine Learning”, 2022.
• (Background) Mehran Sahami, “A Very Brief Introduction to Probability and Machine Learning with the Perceptron Algorithm”, 2021.
• (Explore) Karen Hao and Jonathan Stray, “Can you make AI fairer than a judge? Play our courtroom algorithm game”, 2019.
• (Explore) Martin Wattenberg et al., “Attacking discrimination with smarter machine learning”, 2016.
• Sam Corbett-Davies et al., “A computer program used for bail and sentencing decisions was labeled biased against blacks. It’s actually not that clear”, 2016.
• Julia Angwin, “Make Algorithms Accountable”, 2016.
• Julia Angwin et al., “Machine Bias”, 2016.

Discussion Questions

1. Name at least three factors that introduce or contribute to bias in algorithms and machine-learning models used in decision-making. What proposed solutions exist to eliminate or mitigate the factors that lead to such bias?
2. Recall the Machine Bias article. Why did ProPublica argue that the COMPAS algorithm is biased against black defendants? How did Northpointe, the developer of the algorithm, respond to this criticism according to the Washington Post’s article?

---