Fairness and Machine Learning
Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind (2023)
One-Sentence Thesis
The book 'Fairness and Machine Learning' explores the limitations and opportunities of fairness in machine learning, discussing various aspects such as demographic disparities, statistical non-discrimination criteria, and causal models.
Argument Outline
- 1Introduction to fairness and machine learning
- 2Demographic disparities and the machine learning loop
- 3Statistical non-discrimination criteria and causal models
- 4Relative notions of fairness and systematic relative disadvantage
- 5Understanding United States anti-discrimination law and its limitations
- 6Testing discrimination in practice and algorithmic systems
- 7A broader view of discrimination and structural interventions
Key Distinctions
Demographic disparities vs. statistical non-discrimination criteria
Causal models vs. observational criteria
Relative notions of fairness vs. absolute notions of fairness
Key Terms
Demographic disparities
Statistical non-discrimination criteria
Causal models
Flashcards
14 cardsRelated Questions
In Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind's "Fairness and Machine Learning", social scientists reasons about which of the following?
In Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind's "Fairness and Machine Learning", Sufficiency defines which of the following?
In Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind's "Fairness and Machine Learning", algorithmic interventions applies which of the following?
In Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind's "Fairness and Machine Learning", middle view of equality of opportunity contrasts with which of the following?
Which of the following does Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind contrasts with in "Fairness and Machine Learning"?
In Barocas, Solon, Hardt, Moritz, and Narayanan, Arvind's "Fairness and Machine Learning", Machine learning complicates which of the following?