Causal Inference

CI Courses

Prerequisites: Probability.
Recommended path
  • Take Brady Neal's Causal Inference course and read his free textbook, they're both great! Also take a look at the Mod-U Causal Inference bootcamp by Matt Masten, which is quite complementary as it's more focused on the economics side of CI.
Difficulty Level
Amazing course that covers a lot of ground within Causal Inference from a ML perspective. It does a tremendous job balancing different branches of CI (economics, political science, CS) and tying them back together in the context of ML. The course also brings in some great guest speakers like Susan Athey, Yoshua Bengio and Alberto Abadie.
Over 100 bite-sized videos to understand ideas like counterfactuals, instrumental variables, differences-in-differences and regression discontinuity in a really accessible manner. If any of the concepts in Brady's course seems difficult to grasp, take a look at an analogous Mod-U video first as it's usually more high-level.

Supplementary Resources

The corresponding textbook to the course!
Jonas Peters, Dominik Janzing & Bernhard Schölkopf - Elements of Causal Inference
This book by Peters et al gives the reader a broad overview of causality and some of its connections to ML. 200 pages of well-written content on the cause-effect problem, multivariate causal models, hidden variables, time series etc. Make sure to take a look at this 4-part lecture series by Jonas Peters, which contains a lot of the same topics.
Judea Pearl, Madelyn Glymour & Nicholas Jewell - CI in Statistics
One of the clearest intros to Pearlian Causal Inference. It gradually eases the reader into the world of interventions and counterfactuals from a statistics perspective.
MLSS Tübingen - Causality Tutorial
This ML Summer School (MLSS) has a neat tutorial on causality. Part II by Stefan Bauer is particularly interesting because it goes over causality in ML and some recent developments in causal representation learning.
MLSS Africa - Causal Inference Talks
Beyond a collection of other great talks, this MLSS has recorded Causal Discovery lectures by Bernhard Schölkopf and a more introductory tutorial on Causal Inference in Everyday ML by Ferenc Huszár who also has a corresponding blog post series on the topic, which is one of the most accessible introductions to CI.