Deep Learning and Reinforcement Learning Week 5
Course Name: Deep Learning and Reinforcement Learning
Course Link: Deep Learning and Reinforcement Learning
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Practice: Transfer Learning
Q1. The main idea of transfer learning of a neural network is:
To keep the early layers of a pre-trained network and re-train the later layers for a specific application.
To use the early layers to capture features that are more particular to the specific data you are trying to classify.
To train the early layers such that their weights have a higher impact on the final result.
To re-train the early layers for a specific application and transfer it to a different data set
Answer: To keep the early layers of a pre-trained network and re-train the later layers for a specific application.
Q2. In the context of transfer learning, which is a guiding principle of fine tuning?
Fine tuning the hyperparameters of the CNNs
Using data that is similar to the pre-trained network
Adjust the weights of the neural network
Increase the number of later layers iteratively
Answer: Using data that is similar to the pre-trained network
Q3. In the context of transfer learning, what do we call the process in which you only train the last or a few layers instead of all layers of a neural network?
Frozen layers
Frozen weights
Updated learning
Updated layers
Answer: Frozen layers
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Practice: Convolutional Neural Network Architectures
Q1. This concept came as a solution to CNNs in which each layer is turned into branches of convolutions:
Inception
Workload portion
Hebbian Principle
Network Concatenation
Answer: Inception
Q2. Which CNN Architecture is considered the flash point for modern Deep Learning?
AlexNet
VGG
Inception
ResNet
LeNet
Answer: AlexNet
Q3. Which CNN Architecture can be described as a “simplified, deeper LeNet” in which the more layers, the better?
Deep Lenet
AlexNet
VGG
Inception
ResNet
Answer: VGG
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Q4. Which CNN Architecture is the precursor of using convolutions to obtain better features and was first used to solve the MNIST data set?
AlexNet
VGG
Inception
ResNet
LeNet
Answer: LeNet
Q5. The motivation behind this CNN Architecture was to solve the inability of deep neural networks to fit or overfit the training data better when adding layers.
LeNet
AlexNet
VGG
Inception
ResNet
Answer: ResNet
Q6. This CNN Architecture keeps passing both the initial unchanged information and the transformed information to the next layer.
LeNet
AlexNet
VGG
Inception
ResNet
Answer: ResNet
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Q7. Which activation function was notably used in AlexNet and contributed to its success?
ReLU (Rectified Linear Unit)
Sigmoid
Tanh
Leaky ReLU
Answer: ReLU (Rectified Linear Unit)
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Practice: Regularization
Q1. Which regularization technique can shrink the coefficients of the less important features to zero?
L1
Dropout
L2
Batch Normalization
Answer: L1
Q2. (True/False) Batch Normalization tackles the internal covariate shift issue by always normalizing the input signals, thus accelerating the training of deep neural nets and increasing the generalization power of the networks.
True
False
Answer: True
Q3. Regularization is used to mitigate which issue in model training?
Both underfitting and overfitting
High bias and low variance
Overfitting
Underfitting
Answer: Overfitting
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Final Quiz
Q1. (True/False) In Keras, the Dropout layer has an argument called rate, which is a probability that represents how often we want to invoke the layer in the training.
True
False
Answer: False
Q2. What is a benefit of applying transfer learning to neural networks?
Train early layers for specific applications and generalize that with later pre-trained layers.
Save early layers for generalization before re-training later layers for specific applications.
Easily adjust weights of early layers to reduce training time.
Place heavy focus on training layers that generalize the model.
Answer: Save early layers for generalization before re-training later layers for specific applications.
Q3. By setting ` layer.trainable = False` for certain layers in a neural network, we____
exclude the layers during training because they should be discarded
freeze the layers such thattheir weights change synchronously during training.
set the layers’ weights to zero
freeze the layers such that their weights don’t update during training.
Answer: freeze the layers such that their weights don’t update during training.
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Q4. Which option correctly orders the steps of implementing transfer learning?
1. Freeze the early layers of the pre-trained model.
2. Improve the model by fine-tuning.
3. Train the model with a new output layer in place.
4. Select a pre-trained model as the base of our training.
3, 2, 4, 1
4, 2, 3, 1
3, 1, 2, 4
4, 1, 3, 2
Answer: 4, 1, 3, 2
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Q5. Given a 100×100 pixels RGB image, there are _____ features.
300
100
10000
30000
Answer: 30000
Q6. Before a CNN is ready for classifying images, what layer must we add as the last?
Dense layer with the number of units corresponding to (number of classes*input size)
Dense layer with the number of units corresponding to the number of classes
Flattening layer with the number of units corresponding to the number of classes
Flattening layer with the number of units corresponding to (number of classes*input size)
Answer: Dense layer with the number of units corresponding to the number of classes
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
Q7. In a CNN, the depth of a layer corresponds to the number of:
color channels
input layers
filters applied
channel-filter combinations
Answer: filters applied
These are answers of Deep Learning and Reinforcement Learning Week 5 Quiz
More Weeks of this course: Click Here
More Coursera Courses: http://progiez.com/coursera
