Given a training dataset, does Logistic Regression learn a locally optimal error and the best decision boundary, or does it stop improving as soon as it finds a hyperplane that separates the classes?...

Extracted text: Given a training dataset, does Logistic Regression learn a locally optimal error and the best decision boundary, or does it stop improving as soon as it finds a hyperplane that separates the classes? None of the others. Best boundary O First separating hyperplane A near optimal boundary


Extracted text: In the MNIST digits dataset, there are ten classes {0.9}. How many dimensions would a Logistic Regression decision boundary be between two of its classes? (Check all that apply.) 11 Infinite Decision boundaries do not have dimensions None of the others 3 9 v 10 2.