info@avatto.com
+91-9920808017
31. In supervised learning, training set of data includes
Input
Output
Both input and output
None
Your email address will not be published. Required fields are marked *
Report
Name
Email
Website
Save my name, email, and website in this browser for the next time I comment.
Comment
32. Consider the feedforward neural network in the figure shown below which has one in the layer, two hidden layers, and one output layer. The input x to this network ∈ R3 and number of neurons in the two hidden layers and the output layer are 4, 3, 5 respectively. Each layer is fully connected to the next layer, i.e., there is a weight connecting every neuron in layer i to every neuron in layer i + 1. further every neuron in the hidden output layers also has a bias connected to it. What is the total number of parameters in this feedforward neural network?
60
51
39
44
33. Consider the function f(θ) = f(x, y, z) = x2 + y2 + z 2 - 8. Suppose you start with θ0 = {1, -1, 1} and rund one step of gradient descent with the learning rate η = 1. What will be the updated value of θ
[1, -1, 1]
[1, -1, -1]
[-1, -1, 1]
[-1, 1, -1]
34. Consider the vector a = [1.2,-2.5,2.4,3]. What will the following line of code do? import numpy as np a = np.asarray ([1.2, -2.5, 2.4, 3]) a = np.exp(a) a = a/np.sum(a) 1. It will compute the softmax of a 2. It will compute the element-wise sigmoid of a 3. It will compute the e-th power of each element of a
1 only
2 only
1 and 3 only
3 only
35. Consider a box which contains 100 balls of which 30 are red, 50 are green and 20 blue. Your friend peeps into the box and estimates the number of red, green and blue balls as 50, 25, 25. What is the cross entropy between the true distribution over colors in the box and the distribution predicted by your friend?
1.5
1.0
1.7
2.3
Login with Facebook
Login with Google
Forgot your password?
Lost your password? Please enter your email address. You will receive mail with link to set new password.
Back to login
