If Data Is Linearly Separable Weights Go To Infinity. The two dimensional data above are clearly linearly separable. In fact, an infinite number of straight lines can be drawn to separate the blue balls.

The two dimensional data above are clearly linearly separable. It doesn’t matter if more than one such. In fact, an infinite number of straight lines can be drawn to separate the blue balls.

The Recipe To Check For Linear Separability Is:

The two dimensional data above are clearly linearly separable. In fact, an infinite number of straight lines can be drawn to separate the blue balls. It doesn’t matter if more than one such.

When Data Is Linearly Separable.

The point is that if you train on linearly separable data, you will keep seeing your loss go down forever but really you're just inflating the norm of the weight vector and not actually learning.