A research-grade, 6-week masterclass in Causal Inference and Causal ML from first principles. Rebuilds d-separation oracles, propensity score IRLS engines, doubly-robust AIPW estimators, Cross-Fitting Double Machine Learning (DML), and honest causal forests from scratch in pure NumPy. Fully verified against causal truth
python dml artificial-intelligence causal-inference instrumental-variables dags cate double-machine-learning meta-learners directed-acyclic-graphs conditional-average-treatment-effects inverse-probability-weighting causal-forests potential-outcomes selection-bias-decomposition neyman-orthogonality randomization-mechanics back-door-criterion front-door-criterion regression-discontinuity-design
-
Updated
May 30, 2026 - Jupyter Notebook