#dual-representation

Question 1: Single choice

mediumAI-checkedNearest Neighbors & Kernel Methods

The perceptron updates $\theta \leftarrow \theta + y_t\phi(x_t)$ whenever $y_t\theta^\top\phi(x_t) \le 0$ , starting from $\theta = 0$ . Its kernel form keeps the discriminant as $f(x) = \sum_{i=1}^{n} w_i\, y_i\, K(x_i, x)$ . What is $w_i$ ?

A Lagrange multiplier between 0 and a constant $C$ 1
How many updates (mistakes) were made on $x_i$ so far2
1 if $x_i$ is a support vector and 0 otherwise3
The kernel value $K(x_i, x_i)$ 4

#dual-representation

Timed test