Course Overview
Applied Deep LearningNeural Networks & Backpropagation Checkpoint
0% Done
Knowledge Checkpoint Quiz15 minutes
Neural Networks & Backpropagation Checkpoint
Lesson Summary: Verify your understanding of computation graphs, activation functions, and gradient updates.
Checkpoint Score
0 / 3 Correct
Novice(0 XP)
Q1.What is the primary function of the activation function in a deep neural network?
mcqTo introduce non-linearity so the network can approximate non-linear functions
To normalize the input vectors to standard normal distribution
To speed up GPU matrix multiplications during the forward pass
To prevent integer overflow in float64 arrays
Q2.Which of the following activation functions suffers from the vanishing gradient problem when inputs are very large positive or negative numbers?
mcqSigmoid
ReLU (Rectified Linear Unit)
Leaky ReLU
ELU
Q3.Which techniques help prevent vanishing gradients in deep architectures?
msqResidual skip connections (ResNet)
Batch Normalization / Layer Normalization
Using ReLU or GELU activations
Increasing the learning rate to 10.0