Fairness and Accountability Design Needs for Algorithmic Support in High-Stakes Public Sector Decision-Making

This course aims to explore the design needs for fairness and accountability in algorithmic support for high-stakes public sector decision-making. The learning outcomes include understanding challenges in imbuing public values into machine learning practices, identifying disconnects between organizational realities and research findings, and recognizing design opportunities to enhance transparency and usability in decision-making processes. The course covers topics such as automating decisions, irregular data, political challenges, and ethical considerations. The teaching method involves a combination of interviews, case studies, and theoretical discussions. This course is intended for public sector machine learning practitioners, policymakers, researchers, and anyone interested in the ethical implications of algorithmic decision-making in public services.