CUHK-STAT3009 | Schedule

Date	Lecture	Readings	Logistics
Thu 09/10	Lecture #1 Course Overview & Software Setup Open notebook Slides	Python NumPy Tutorial - CS231n Stanford Python Tutorial VS Code & Google Colab Setup: Install & Quick Start · Step-by-step Guide
Thu 09/17	Lecture #2 Baseline Methods for Rating Prediction Open notebook Slides PDF	Warm-up Pandas from a URL In-class review pandas: Summary statistics and groupby NumPy: Integer and Boolean array indexing The Netflix Prize (Bennett & Lanning, 2007) (Sections 1–3: background, evaluation, and data splits) Optional: Recommender Systems Datasets (UCSD)
Thu 09/24	Lecture #3 Machine Learning I: Models and Estimators Open notebook Slides PDF	Warm-up Core pandas + NumPy review 15–18 min scikit-learn: Getting started (fit, predict, and evaluate) scikit-learn: LinearRegression scikit-learn glossary: Parameters and hyperparameters scikit-learn: Mean squared error Optional: Developing scikit-learn estimators	HW1 released Open HW1 in Colab Due: Thu 10/08 before lecture
Thu 10/08	Lecture #4 Machine Learning II: Generalization and Model Selection Open notebook Slides	Warm-up Models, estimators, and predictions 15–18 min scikit-learn: Cross-validation and data splitting (introduction, K-fold, and group-wise splits) scikit-learn: Ridge regression (objective and `alpha`) scikit-learn: GridSearchCV (parameter grid, scoring, and refit) Optional: The BellKor Solution to the Netflix Grand Prize (Sections II–III: rating splits, RMSE, and baseline validation)	HW1 due before this lecture
Thu 10/15	Lecture #5 Matrix Factorization for Recommendation Open notebook Slides Notes	Warm-up Build, fit, predict, select 18–20 min Google: Collaborative filtering and learned user/item profiles Koren, Bell & Volinsky: Matrix Factorization Techniques for Recommender Systems (pp. 43–46: latent factors, bias terms, and ALS) Optional: Simon Funk: Netflix Update — Try This at Home (an intuitive account of learned factors) Optional, more technical: Hastie et al.: Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares (Section 5: ALS and ridge subproblems)
Thu 10/22	Lecture #6 From EDA to Recommender: CUHK Canteen Explorer Open notebook	Warm-up MF workflow — fit, predict, select 18–20 min Seaborn: Overview of plotting functions Seaborn: Visualizing distributions Seaborn: Visualizing statistical relationships Optional: pandas: Visualization
Thu 10/29	Mid-term Jupyter Notebook Exercise Baseline models and matrix factorization models