There will be eight homework assignments. Odd numbered assignments will be theoretical in nature. Even numbered assignments will be computational in nature. Assignments must be uploaded to Gradescope. You may type your responses or handwrite them. You can upload a pdf or individual photographs of each page of hand written work. For computational assignments, you must write your solutions as a Jupyter notebook and include a pdf of the Jupyter notebook as part of your submission.
Midterm I and Midterm II will be take-home exams. They will be posted at noon on exam day and will be due by noon on the following day. You may type your responses or handwrite them. You can upload a pdf or individual photographs of each page of handwritten work. The exams will not involve writing code. You are expected to work individually on the exams. You may consult books and/or the internet, but you may not consult other people.
The Project will be on a topic of your choosing. You will be able to choose a project of one of the following forms: (a) Select a dataset and a machine learning task and solve the machine learning task in at least two different ways; (b) Select a paper about machine learning and implement the described method and at least one baseline.
| Day | Date | HW | Class Will Be On: | Readings |
|---|---|---|---|---|
| 1 | W 9/8 | Introduction to Machine Learning (Notes) | Chapter 1 of Mattmann. (Also relevant are Chapters 1-4 of Zero to AI) | |
| 2 | M 9/13 | Review: Linear Algebra, Calculus (Notes) | Mathematics for Machine Learning, Sections: 3.1, 3.2, 3.4, 3.6, 3.17.1, 4.1, 4.2, 4.3, 4.6 | |
| 3 | W 9/15 | Review: Probability, Statistics (Notes) | Mathematics for Machine Learning, Sections: 5 (Except 5.8.2 and 5.9) | |
| 4 | M 9/20 | Linear Regression (notes unannotated, annotated) | Hastie et al. Section 3.2 "Linear Regression Models and Least Squares" (up until but not including equation (3.8) ). | |
| 5 | W 9/22 | HW 1 (theory) Due (pdf, tex) | Introduction to TensorFlow (notes) | Mattmann Chapter 3 |
| 6 | M 9/27 | Classification and Logistic Regression (notes unannotated, annotated) | Mattmann Chapter 5 | |
| 7 | W 9/29 | HW 2 (computation) Due (pdf, tex). Revisions Due 10/31 | Logistic Regression (notes unannotated, annotated) | Mattmann Chapter 5 |
| 8 | M 10/4 | Statistical Learning Framework (unannotated, annotated) | Chapter 2 of Shalev-Shwartz and Ben-David | |
| 9 | W 10/6 | HW 3 (theory) Due (pdf, tex) | Statistical Learning Framework(unannotated, annotated) | |
| - | M 10/11 | No Class - Indigenous Peoples Day | ||
| 10 | W 10/13 | HW 4 (computation) Due (pdf, tex). Revisions due 11/12. | Bias Variance Tradeoff (notes unannotated, annotated) | |
| 11 | M 10/18 | Ridge Regression (notes unannotated, annotated) | Hastie Section 3.4.1 | |
| 12 | W 10/20 | HW 5 (theory) Due (pdf, tex) | MAP Estimation, Model Validation (notes unannotated, annotated) | Hastie Sections 3.4.1 and 7.10 |
| 13 | M 10/25 | Review for Midterm (unannotated, annotated) | ||
| 14 | W 10/27 | Midterm I (pdf, tex) | Study Guide and Practice Problems (solutions) | |
| 15 | M 11/1 | Cross Validation, Nearest Neighbor Methods (notes unannotated, annotated) | Hastie 2.3.2, 2.3.3, 13.3 | |
| 16 | W 11/3 | Gradient Descent, Stochastic Gradient Descent (notes unannotated, annotated) | ||
| 17 | M 11/8 | Convex Optimization, Convergence of Gradient Descent (notes unannotated, annotated) | ||
| 18 | W 11/10 | HW 6 (computational) Due (pdf, tex). Revisions due 12/10. | Clustering, k-means (notes unannotated, annotated) | Bishop 9.1 |
| 19 | M 11/15 | Mixtures of Gaussians, EM Algorithms (notes unannotated, annotated) | Bishop 9.2, 9.3 | |
| 20 | W 11/17 | HW 7 (theory) Due (pdf, tex) | Principal Component Analysis (notes unannotated, annotated) | Bishop Section 12.1 |
| 21 | M 11/22 | Project Descriptions Due (pdf, tex) | Work on Projects | |
| - | W 11/24 | Thanksgiving | ||
| 22 | M 11/29 | Midterm Review (notes annotated) | ||
| 23 | W 12/1 | Midterm II (pdf, tex) | Study Guide and Practice Problems (solutions) | |
| 24 | M 12/6 | Guest Speaker from Industry | ||
| 25 | W 12/8 | Neural Networks - Architectural Elements (notes unannotated) | ||
| 26 | M 12/13 | Neural Networks - Architectures (notes unannotated) | ||
| 27 | W 12/15 | Final Project Due | Neural Networks |