Machine Learning 102 - Unsupervised Learning

Overview of the course

Machine Learning 102 surveys unsupervised learning and fault detection.

The class begins at the level of undergraduate probability and statistics and introduces a broad array of unsupervised learning techniques. The classes give participants a working knowledge of these techniques and leave them prepared to apply them to real problems. To get the most out of the class, participants will need to work through the homework assignments.

Prerequisites

This class assumes a moderate level of computer programming proficiency. We will use R (the open source statistics language) for the homework and for the examples in class. We will cover some of the basics of R and do not assume any prior knowledge of R. You can find references to how to use R on the class website and we will give out sample code during classes that will help get you started.

You'll need some general beginner-level background in probability, calculus, linear algebra and vector calculus. We will cover most of what is required during the lectures. The appendices in the back of the Tan text are more than sufficient level for this class.

Course Schedule

10/22/2011 – Hierarchical, Density, & K - means Clustering

10/29/2011 – Expectation Maximization Algorithms & Discriminant Analysis

11/5/2011 - Anomaly Detection One Class SVM & One-D Statistical Methods

11/12/2011 – Outliers, Extreme Values, & Convex Hull

11/19/2011 – Class Presentations & Projects

Classes run from 9:00 am until 1:00 pm for 5 Saturdays