Statistics 315a: Modern Statistical Learning

John Duchi, Stanford University, Winter 2024

Syllabus

As on the home page, this class has three main foci:

prediction methods based on linear models
inferential goals, including parameter recovery, predictive validity, and validation
advanced modeling, including kernel methods and model selection

Lecture	Date	Topics	Reading(s)
1	8 Jan	Overview of supervised learning	ESL Chapters 1, 2
2	10 Jan	Supervised learning: regression and kNN	ESL Chapters 1, 2
3	17 Jan	Least squares, potential outcomes	ESL Chapters 3.1–3.2, CIS Chapter 1
4	22 Jan	Potential outcomes, loss functions	ESL Chapter 4
5	24 Jan	Generative prediction models, trees	ESL Chapter 4
6	29 Jan	Boosting	ESL Chapters 9.1–9.2, 10.1–10.11
7	31 Jan	Cross validation, begin parameter inference	ESL 7.1–7.5, 7.10–7.12
8	5 Feb	Parameter inference	asymptotics
9	7 Feb	Parameter inference	asymptotics
10	12 Feb	Conformal prediction and validation	AB
11	14 Feb	Validation methods	AB, ESL 7.10–7.12
12	21 Feb	Basis expansions and kernel methods	ESL Chapters 5, 6

Reading key

ESL: The Elements of Statistical Learning; Hastie, Tibshirani, and Freedman
CIS: Causal Inference for Statistics, Social, and Biomedical Sciences; Imbens and Rubin.
AB: Angelopoulos and Bates, A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification