Short Description

Introduction to supervised machine learning, with a focus on classification. Decision trees, logistic regression, and basic machine learning concepts such as generalization error and overfitting.

Learning Outcomes

By the end of the course, students are expected to be able to:

  1. Explain, with examples, the key differences between an unsupervised learning problem and a supervised learning problem, as well as the concept of training and test data.
  2. Construct and apply a decision tree classification model, and explain how the concept of generalization error is tied to the depth of a decision tree.
  3. Apply logistic regression to classification, and explain the key differences between regression and classification models.
  4. Apply regularization in the context of logistic regression.
  5. Build a k-th nearest-neighbor (kNN) classifier; compare and contrast parametric and non-parametric classification models.
  6. Deploy support vector machine (SVM) classifiers, and explain how kernel functions are used in such classifiers.
  7. Avoid the pitfalls of overfitting and reusing test sets.

Reference Material

  • Russell, Stuart, and Peter Norvig. Artificial intelligence: a modern approach. Third Edition. 2010.
  • David Poole and Alan Mackwordth. Artificial Intelligence: foundations of computational agents. 2010. (free online http://artint.info/)

Instructor (2016-2017)

Note: information on this page is preliminary and subject to change.