Overview
This course aims to teach learners how to assess fairness in machine learning models, specifically focusing on the concept of recourse. By the end of the course, students will be able to audit linear classification models in terms of recourse, enabling individuals to understand and potentially change the decisions made by these models. The course covers topics such as immutable variables, the right to an explanation, ethical considerations for data scientists, optimization approaches, and addressing demographic differences. The teaching method involves a video presentation, and the course is designed for data scientists, decision-makers, and journalists interested in understanding and evaluating the fairness of machine learning models.
Syllabus
Introduction
What is ODSC
What does this mean
A growing appetite to understand ML
What is this work
Example of a data scientist
Recourse
Immutable variables
Why does that matter
Right to an Explanation
Ethical Data Scientists
Explanations
Optimization Approach
Optimization Equation
Cost
Optimization
Problems
Demographic Differences
Confounding
Conclusion
Taught by
Open Data Science