Relating the Binary and Multi-class Perceptron (handwritten notes)

Another off-the-cuff handwritten notes for students learning about binary and multi-class linear classifiers. I go over how the multi-class perceptron generalizes the binary perceptron and you can slightly hear my parakeet chirping in the background.

[@utkarshkathuria2931]

what is the difference between Wy and Wf(x) at 4:45?

[@berty38]

y is the true label and f(x) is the label estimated by the current model

[@utkarshkathuria2931]

@@berty38 so basically how did you assume W+transpose X is greater or less than W-transposeX. What if I take sign ( W-transposeX - W+transpose X)?

[@maged4087]

what is the y ?