Courses

This course serves as a modern, applied introduction to machine learning. Students will learn how to evaluate machine learning models and learn specific methods in supervised and unsupervised learning, including regression, ensembles, and neural networks. Other frontier topics with social science relevance will be presented. Topics will be of interest to researchers who are interested in prediction, causal inference, text analysis, and more. Students may use Python, R, or any coding
language that requires only free software. Lectures and lecture notes will only include Python. Students should have prior experience with regression models, be comfortable with matrix algebra notation, and have experience with basic coding (R or Python, ideally).

Learning goals:
• Understand common machine learning models and be able to implement them.
• Gain familiarity with the use of machine learning models in modern social science research.
• Be able to identify research settings where different models might be appropriate.
• Be able to read and interpret technical research papers, extracting the key methodological
choices, assumptions, and results.