Answer Happy

1 Problem 1 (15 points) Suppose we have trained a logistic regression model for several iterations on a tiny dataset with two features (see Table 1), and the resulting parameters are W₁ = -0.25, W₂ = -1.01 (slope) and b = 0.41 (intercept). (1) Compute p(y=0|1, 2) for all the samples using the parameters above. (5 points) (2) Compute the log likelihood value. (5 points) (3) How can you transform the data such that they can be linearly separated? (5 points) No. I1 1 2 -1.0 3 -0.4 4 0.5 X2 1.4 0.9 Y 0 -1.3 1 0.5 0 1.0 1 Table 1: Tiny dataset