USMA MA289: Mathematics of AI

Unsupervised Learning: Clustering & Dimensionality Reduction

Overview

Unsupervised learning focuses on discovering structure in data without labeled outcomes. In this block we examine clustering methods, dimensionality reduction techniques, and feature extraction.

Topics Covered

Resources

Key Topics, Meeting Notes / Lesson Objectives, and Assignments

Date Key Topics Meeting Notes and Lesson Objectives Assignments
07 Apr 2026 Lesson 12: Unsupervised Learning 1 Overview of clustering and PCA. Understand when unsupervised methods are appropriate and how they differ from supervised approaches. ISLP 503-534; MLP Mod 4
14 Apr 2026 Lesson 13: Unsupervised Learning 2 Compare clustering algorithms (K-Means, Hierarchical, DBSCAN) and discuss distance metrics, cluster assumptions, and evaluation methods. Run K-Means on a dataset of your choice and visualize results. Experiment with different numbers of clusters.
21 Apr 2026 Lesson 14: Project 3 Practical Unsupervised Model 1 Understand variance maximization (PCA) and nonlinear manifold learning (t-SNE). Peform Unsupervised learning on selected dataset. Complete Report
28 Apr 2026 Lesson 15: Project 3 Practical Unsupervised Model 2 Turn in and Present unsupervised learning project 3. Demonstrate unsupervised learning on selected dataset in front of class.
5 May 2026 Lesson 16: Regularizatoin Review regularization techniques and explore uses. ISLP 229-252; MLP Mod 5