A short, hands-on refresher on machine learning fundamentals through the lens of algorithmic discrimination in ride-sharing platforms (Uber, Lyft). Each module pairs an ML concept with a real discrimination scenario; all exercises are in R.
| # | Module | Status |
|---|---|---|
| 1 | The Learning Problem — bias-variance, train/val/test, statistical vs social bias | ✓ done |
| 2 | Linear Models — linear & logistic regression, regularization, proxy discrimination | ✓ done |
| 3 | Model Evaluation & Selection — confusion matrix, ROC/AUC, calibration, per-group audits | ✓ done |
| 4 | Tree-Based Methods — decision trees, random forests, gradient boosting, SHAP | ✓ done |
| 5 | Unsupervised Learning | upcoming |
| 6 | Neural Networks Foundations | upcoming |
| 7 | Fairness Frameworks & Metrics — demographic parity, equalized odds, predictive parity, impossibility theorem | ✓ done |
| 8 | Auditing & Interpretability | upcoming |