The Perceptron Algorithm Primal Form

1
The Perceptron Algorithm (Primal Form)
Repeat
until no mistakes made within the for loop return
?
. What is
2
(No Transcript)
3
The Perceptron Algorithm
( STOP in Finite Steps )
Theorem 2.3 (Novikoff)
be a non-trivial training set, and let
Let
4
The Perceptron Algorithm (Dual Form)
Given a linearly separable training set
and
Repeat
until no mistakes made within the for loop return
5
What We Got in the Dual Form Perceptron
Algorithm?

• The number of updates equals

6
The Margin Slack Variable of
with respect to
7
(No Transcript)
8
Bound of Mistakes of a for loop
for the Perceptron Algorithm
9
Fishers Linear Discriminator
10
Multi-class Classification

• Equivalent to our previous rule

• For multi-class (one against rest)