Suppose you have the following training set, and fit a logistic regression classifier : ho(x) =...

 

Transcribed Image Text:

▪ Suppose you have the following training set, and fit a logistic regression classifier : ho(x) = g(00+011+0₂x2) O O O Which of the following are true? Check all that apply. a) Adding polynomial features e.g., ho(x) = g(00+011+0₂x2 +03x² +04x1x2 +05.x²) could increase how well we can fit the training data. b) At the optimal value of 0 (e.g., found by fminunc), we will have J(0) ≥ 0. c) Adding polynomial features e.g.,he(x) = g(00+0₁x1 +02x2 +03x² +04x1x2 +05x²) x₁ 1 2 3 would increase J(0) because we are now summing over more terms. d) If we train gradient descent for enough iterations, for some examples (¹) in the training set it is possible to obtain he((¹)) > 1 X2 0.5 1.5 1 1 y 0 0 0 ▪ Suppose you have the following training set, and fit a logistic regression classifier : ho(x) = g(00+011+0₂x2) O O O Which of the following are true? Check all that apply. a) Adding polynomial features e.g., ho(x) = g(00+011+0₂x2 +03x² +04x1x2 +05.x²) could increase how well we can fit the training data. b) At the optimal value of 0 (e.g., found by fminunc), we will have J(0) ≥ 0. c) Adding polynomial features e.g.,he(x) = g(00+0₁x1 +02x2 +03x² +04x1x2 +05x²) x₁ 1 2 3 would increase J(0) because we are now summing over more terms. d) If we train gradient descent for enough iterations, for some examples (¹) in the training set it is possible to obtain he((¹)) > 1 X2 0.5 1.5 1 1 y 0 0 0

Answer rating: 100% (QA)

a TRUE Adding polynomial features increses how well we fit the da... View the full answer