Algorithmic Fairness

Catalog description

The topic is how software, particularly AI/machine learning software, can unintentionally contribute to bias and discrimination against certain groups of people, based on race, ethnicity, gender, religion, sexual orientation, economic advantages and other factors, and what can be done to remedy this problem. The course will cover case studies of actual algorithmic bias; alternative types and characterizations of algorithmic bias and fairness; detection and measurement of algorithmic bias; causes of algorithmic bias; and technical and nontechnical remedies and their limitations. Prereq: CSDS 391 or CSDS 440 .