Which is shown in the graphic below. The two dimensions are X and Y, and the shade of each dot represents what class it is. You want to classify this data accurately using a linear algorithm. To do this you need to add a synthetic feature. What should the value of that feature be?

5 thoughts on “Which is shown in the graphic below. The two dimensions are X and Y, and the shade of each dot represents what class it is. You want to classify this data accurately using a linear algorithm. To do this you need to add a synthetic feature. What should the value of that feature be?”

1. It’s D, top dots are black, no grey

   1

   Reply

   1. D
      top dots are black & “To do this you need to ADD a synthetic feature.” -> in model are 3 feature: x, y, cos(x)

      Reply

2. By the way….it’s A

   1

   Reply

3. can’t be that (cos(x). the 2 variables that make a circle in http://playground.tensorflow.org are x1^2 and x2^2.
   Sin(x) or cos(x) whould just make horizontal stripes.

   To do this you’d use those 2 variables, learning rate 0,3 for example, classification type, no regularization needed and any activation function will work fine.

   Reply

4. A is the equation of a circle which looks like a good separation for the black vs grey dots.

   2

   Reply