Explore a comprehensive tutorial from FAT* 2018 that delves into 21 different definitions of fairness in machine learning and their political implications. Led by Arvind Narayanan from Princeton University, this 55-minute session unpacks the complex relationship between mathematical fairness criteria and social understandings of justice. Examine the trade-offs between various fairness notions, such as individual vs. group fairness and statistical parity vs. error-rate equality. Gain insights into how technical discussions about fairness definitions intersect with important normative questions. Learn why the proliferation of fairness definitions should be embraced rather than avoided, and understand the limitations of seeking a single, universal definition. Connect technical observations to philosophical theories of justice, making this tutorial valuable for computer scientists, policymakers, ethicists, and domain experts alike.
Overview
Syllabus
FAT* 2018 Translation Tutorial: 21 Definitions of Fairness and Their Politics
Taught by
ACM FAccT Conference
Related Courses
-
Translation Tutorial - A History of Quantitative Fairness in Testing
-
Fairness in Machine Learning - Lessons from Political Philosophy
-
Toward a Theory of Race for Fairness in Machine Learning
-
Challenges of Incorporating Algorithmic Fairness into Industry Practice
-
Translation Tutorial - A Shared Lexicon for Research in Human-Centered Software Systems
-
Decoupled Classifiers for Group-Fair and Efficient Machine Learning
