Questions and Comments for: Bayes Rule to Minimize Risk

A slecture by Andy Park


Please leave me comment below if you have any questions, if you notice any errors or if you would like to discuss a topic further.


Write comments here


The slecture is reviewed by Yu Wang.

In the section revisit Bayes rule/classifier, the author reviewed the basic concept of Bayes rule by illustrating a simple two-fish example. Step by step, concepts like likelihood, posterior and discriminant function are introduced with graphs and numerical deviration. Finally, likelihood ratio test is associated with Bayes rule.

In the second section The concept of Risk/Cost Minimization, the author extended what we learned in the class to more general sense. He first argued that cancer patients who are classified as non-cancer ones are much riskier than the opposite. Then he defined different risk constant, which later is proved to affect decision boundaries. Finally, the author manipulated the risk constants to show us the relationship between MAP and ML estimator.

The third and fourth section basically discusses about loss matrix and Bayes risk with extensive and well explained examples.

This video slecture is pretty much well-organized and well-explained. What impressed me most after watching 50min video is that the author managed to explain everything with tons of formulas and graphs fluently. I did learn a lot from those delicately picked examples, since they are very easy to understand. The only thing I will recommend is that trying to make the materials more compact, aka make video shorter. :)




Back to Bayes Rule to Minimize Risk

Alumni Liaison

Abstract algebra continues the conceptual developments of linear algebra, on an even grander scale.

Dr. Paul Garrett