This repository contains lab materials from the course "Introduction to Causal Inference in Machine Learning" from Spring 2024 at New York University. You can check out the syllabus here.
The course was taught by Kyunghyun Cho, and the lab sessions were prepared and led by three PhD students; Taro Makino, Daniel Jiwoong Im and Divyam Madaan.
The lecture note can be found here, and this repository contains the following lab materials:
- Lab 1: structural causal models and conditional distributions by Taro Makino
- Lab 2: the difference between conditional and interventional distributions by Taro Makino
- Lab 3: randomized controlled trials by Daniel Im
- Lab 4: outcome maximization by Daniel Im
- Lab 5: matching and inverse probability weighting by Daniel Im
- Lab 6: instrument variables and two-stage linear regression by Divyam Madaan
- Lab 7: a tutorial on Numpyro by Divyam Madaan
- Lab 8: double machine learning by Divyam Madaan
- Lab 9: the principle of invariance by Taro Makino
- Lab 10: colliders and the phenomenon of explaining away by Divyam Madaan
These labs were run over Google Colab originally, but in order to ensure the reproducibility and to improve the usability, this lab materials are provided as a lightning studio. I recommend you to use the lightning studio to try out these lab sessions yourselves.