(Solved): 10. What happens if the learning rate is too large in Gradient Descent algorithm? a) It makes it to ...

10. What happens if the learning rate is too large in Gradient Descent algorithm? a) It makes it too slow. b) It may get stuck in a local minimum. c) It overshoots the global minimum. d) All the above can happen. 11. Suppose there are 2 values for an attribute with occurrence probability of 0.3 and 0.7 . What is the entropy of the attribute rounded to 2 digits after the decimal point? a) 0.55 b) 0.62 c) 0.75 d) 0.88 12. Which activation function is best to use in the output layer of an ANN model that performs multi-class classification? a) Sigmoid b) SoftMax c) Linear d) Tanh