Practice Quiz
Deep Neural Networks
Menu
Submit your assignment
Try again
Receive grade
Your grade
100%
We keep your highest score
View Feedback
Like
Dislike
Report an issue
Deep Neural Networks
Practice Quiz • 30 min • 10 total points
Congratulations! You passed!
Grade received
100%
To pass
80% or higher
1.
Which of the following is NOT true about activation functions?
1 / 1 point
Correct
2.
What is NOT true about gradient descent?
1 / 1 point
Correct
G
di
t d
t i
l
th d f
ti
i
ti
hi h i
l
Back
English
Go to next item
Activation functions describe non-linear transformation Activation functions are specified by the user when setting up the neural network architectures
Activation functions are learned directly from the data by neural network models. ReLU is able to cope with vanishing gradient problems better than Sigmoid and Tanh. Log-likelihood and likelihood function has the same optimal but log-likelihood is o±en easier to manipulate. Gradient descent is an optimization method for optimizing model parameters
Gradient descent is a specific design method for neural network optimization. Stochastic gradient descent is a variant of the gradient descent method that is popular for neural networks training.