Level: Graduate
Taught: S2025, F2024, F2022
Teaching Fellows: Yuning Liu (S2024), Ruofan Ma (S2024), Siyun Jiang (F2022)
Description: As social science research increasingly leverages massive or non-traditional datasets—spanning images, audio, text, networks, time series, and geospatial records—traditional generalized linear models often struggle when covariates outnumber observations or data resist matrix form. This course surveys the machine learning and AI strategies developed since the 1950s (with rapid advances since the 1990s) to address these challenges, from tree-based methods and neural networks to sequence, network, ensemble, and probabilistic neural models. Through hands-on implementation and critical discussion, we’ll examine how these tools can drive social science inquiry, explore fairness, privacy, and broader societal implications, and equip you to apply, evaluate, and ethically adapt state-of-the-art ML techniques in research and policy contexts.
–