CS 461: Machine Learning
Instructor: Kiri Wagstaff

Reading Questions for Lecture 3

Support Vector Machines (Ch. 10.1-10.4, 10.6, 10.9)
  1. What is discriminant-based classification? How is this different from the nonparametric methods we've seen so far?
  2. What is the difference between a multiple-class problem that is "linearly separable" and one that is "pairwise separable"?
  3. What are the "support vectors" in a support vector machine?
Evaluation (Ch. 14.5, 14.7, 14.9)
  1. When would you use a t-distribution (p. 337) instead of a regular confidence interval (as on p. 335-336)?
  2. How is a contingency table different from a confusion matrix?