PH.140.644 - Statistical Machine Learning
Methods, theory, and applications of statistical machine learning for public health and biomedical research
First Term 2026
Johns Hopkins Bloomberg School of Public Health
Instructor: Yiqun T. Chen
Teaching Assistant: Junwei (Ivy) Sun
Course Meeting Times: Tuesdays and Thursdays, 1:30–2:50 pm
Location: Wolfe Street Building, W2015
Syllabus: Syllabus (PDF)
Course materials: Lecture slides, homework, and data are posted on CoursePlus.
Course Description
This course introduces the statistical and computational foundations of modern machine learning for public health and biomedical research. Topics include regression, classification, resampling methods, model selection and regularization, non-linear methods, trees and ensembles, deep learning, text and other non-tabular data, unsupervised learning, and large language models. The course emphasizes careful model evaluation, scientific interpretation, and responsible use of AI tools. The textbook is An Introduction to Statistical Learning (2nd edition) by James, Witten, Hastie, and Tibshirani.
Practice Quizzes
Before each quiz, students get a shorter practice version with the same style and difficulty. They are posted here for anyone who wants to see how we test understanding of statistical learning without heavy computation.
| Quiz | Topic | Practice Quiz |
|---|---|---|
| 1 | Statistical learning foundations | |
| 2 | Regression and classification | |
| 3 | Classification: model flexibility and KNN | |
| 4 | Model selection, regularization, and evaluation | |
| 5 | Non-linear methods, trees, and ensembles | |
| 6 | Deep learning and non-tabular data |
Schedule
| Week | Tuesday | Thursday |
|---|---|---|
| 1 | Class organization; overview of statistical learning | Overview of statistical learning and modern data science |
| 2 | Regression: recap and new ideas | Classification (Quiz 1) |
| 3 | Classification and evaluation | Regularization: ridge regression and the lasso (Quiz 2) |
| 4 | Cross-validation | Model selection: choosing tuning parameters (Quiz 3) |
| 5 | Non-linear methods I | Non-linear methods II (Quiz 4) |
| 6 | Deep learning | Moving beyond tabular data (Quiz 5) |
| 7 | Unsupervised learning: dimension reduction | Unsupervised learning: clustering |
| 8 | Large language models | Effective applications of ML/AI methods (Quiz 6) |