If in one step, a perceptron is adjusting the parameters according to some data sample

with label

$,$ which of the following statements are correct?

The linear function in the perceptron is defined by the model parameter vector

To simplify the discussion, we will overload the label

with numerical meaning:

marks the "positive" class to which a sample is assigned when

is positive, and

marks the "negative" class.

Now consider the update taking place in an individual step of model training

Group of answer choices

will decrease.

will increase.

will change to a direction where the training error over the training dataset decreases.

will change towards the same direction of

$,$ i$.$e$.$ it will increase if

is $+1$ and decrease if

is $-1$