COGS 118B

Instructor: Jason Fleischer

Quarters: Winter 2024

This course teaches students the basics of unsupervised machine learning. The topics include Bayesian esimation, Maximum Likelihood Estimation (MLE), clustering algorithms (Hierarchical clustering, K-means, Spectral clustering, and DBSCAN), Gaussian Mixture Models (GMMs), the Expectation-Maximization (EM) algorithm, dimensionality reduction (PCA, t-SNE, and UMAP), and autoencoders. The course contains programming assignments where students use Python packages such as NumPy, Pandas, and Scikit-learn to implement these algorithms. The course ends in a final project where students come up with their own unsupervised machine learning problem, experiment with different algorithms on datasets of their choice, and present their findings.