Bayes Decision Theory: My answers to the lecture questions

Moderator: Statistisches Maschinelles Lernen

MFB
Erstie
Erstie
Beiträge: 21
Registriert: 3. Okt 2013 17:10

Bayes Decision Theory: My answers to the lecture questions

Beitrag von MFB »

How can we decide on classifying a query based on simple and general loss
functions?

Pick the class with the lowest loss-score

What does “Bayes optimal” mean?

It means that the probability of misclassification/ the risk has been minimized. See https://www.quora.com/What-is-meant-by- ... l-solution

How to deal with 2 or more classes?

Still pick the class with the highest Maximum-a-posteriori

How to deal with high dimensional feature vectors?

Use multi-variate probability densities

How to incorporate prior knowledge on the class distribution?

By using Bayes-rule, we incorporate the prior probabilities of the classes into our decision.

Zurück zu „Statistisches Maschinelles Lernen“