In this problem, we will investigate the perceptron algorithm with different iteration ordering.
Consider applying the perceptron algorithm through the origin based on a small training set containing three points:
x¹ = [-1, -1], y¹ =1
x² = [ 1, 0], y²=-1
x³ = [-1, 1.5], y³=1
Given that the algorithm starts with θ⁰ = 0, the first point that the algorithm sees is always considered a mistake. The algorithm starts with some data point and then cycles through the data (in order) until it makes no further mistakes.
How many mistakes does the algorithm make until convergence if the algorithm starts with data point x¹?