Course Description: A four week short course presenting the principles behind when, why, and how to apply modern machine learning algorithms. We will discuss a framework for reasoning about when to apply various machine learning techniques, emphasizing questions of overfitting/underfitting, regularization, interpretability, supervised/unsupervised methods, and handling of missing data. The principles behind various algorithms—the why and how of using them—will be discussed, while some mathematical detail underlying the algorithms—including proofs—will not be discussed. Unsupervised machine learning algorithms presented will include k-means clustering, principal component analysis (PCA), and independent component analysis (ICA). Supervised machine learning algorithms presented will include support vector machines (SVM), neural nets, classification and regression trees (CART), boosting, bagging, and random forests. Imputation, the lasso, and cross-validation concepts will also be covered.
Course assumes no prior background in machine learning. Previous exposure to undergraduate-level mathematics (calculus, linear algebra, statistics) and basic programming (R/Matlab/Python) helpful.
Week | Monday | Wednesday | Assignment |
---|---|---|---|
1 | No lecture | No lecture | |
2 | Lecture 1: Overview of Machine Learning | Lecture 2: Linear and Logistic Regression | |
3 | No lecture (MLK Day) | Lecture 3: Regularization and Sparsity | HW1 due Friday 5:00pm |
4 | No lecture | No lecture | HW2 due Friday 5:00pm |
5 | Lecture 4: Cross-validation and Imputation | Lecture 5: Support Vector Machines | |
6 | Lecture 6: Classification and Regression Trees (CART) | Lecture 7: Unsupervised Methods | HW3 due Friday 5:00pm |
7 | No lecture (President's Day) | Lecture 8: Neural Networks | |
8 | No lecture | No lecture | HW4 due Friday 5:00pm |
Homework assignments will appear here as they are assigned. Solutions will be posted after the due date.
Assignments will be submitted through Google Forms and Gradescope. Part 1 of each assignment is a conceptual multiple choice submitted via Google Forms. Parts 2 and 3 contain exercises that will be submitted via Gradescope. You should have received an invite to Gradescope for CME 250. Login via the invite and submit assignments on time. If you have not received an invite, please email me.
You are encouraged to form study groups and discuss homework with other students. Submitted homework assignments must however be written and coded from scratch independently, and must not refer to notes from group discussion. In other words, each student should understand the solutions deeply enough to reproduce them by him/herself. If homework exercises from the reference text have solutions available online or via other sources, do not copy or refer to them when doing the assignments for this course.