Introduction to Computer Vision and Image Processing Week 4

Course Name: Introduction to Computer Vision and Image Processing

Course Link: Introduction to Computer Vision and Image Processing

These are answers of Introduction to Computer Vision and Image Processing Week 4 Quiz

---

Practice Assessment

Q1. Data augmentation techniques
Add extra layers to the neural network
Randomly change the training data
Prevent overfitting

Answer: Randomly change the training data, Prevent overfitting

---

Q2. What is the problem that occurs when a network gets deeper, resulting in a smaller gradient?
Validation
Gradient descent
Vanishing gradient

Answer: Vanishing gradient

---

Q3. Usually, the more input dimensions there are in a neural network, the…
More neurons are required
Less neurons are required
The number of required neurons remains the same

Answer: More neurons are required

---

These are answers of Introduction to Computer Vision and Image Processing Week 4 Quiz

---

Q4. Which of the following are activation functions? Check all that apply.
ReLU
Sigmoid
Logistic function

Answer: ReLU, Sigmoid, Logistic function

---

Q5. For a given input in a neural network, we obtain an output for each neuron for the last layer. The class chosen is according to the index of the neuron with the…
Largest value
Smallest value
Randomly selected value

Answer: Largest value

---

These are answers of Introduction to Computer Vision and Image Processing Week 4 Quiz

---

Graded Quiz

Q1. Layering multiple linear and logistic functions is best described as
Fully Connected Network
Logistic Regression
Non-Linear Network

Answer: Fully Connected Network

---

Q2. Which of the following helps to reduce the number of parameters of an input image and still preserves the important features?
Layer
Flattening
Receptive field
Pooling

Answer: Pooling

---

These are answers of Introduction to Computer Vision and Image Processing Week 4 Quiz

---

Q3. If I add more neurons to my neural network, what may I expect?
A perfect model
Overfitting
Underfitting

Answer: Overfitting

---

Q4. What makes a neural network a deep neural network?
Having one hidden layer
Having no hidden layer
An overfitting model
Having more than one hidden layer

Answer: Having more than one hidden layer

---

Q5. When we apply logistic regression in the context of Neural Networks, what is that called?
SoftMax Function
Activation Function
Linear Function
Decision Function

Answer: Activation Function

---

These are answers of Introduction to Computer Vision and Image Processing Week 4 Quiz

---

More Weeks of this course: Click Here

More Coursera Courses: http://progiez.com/coursera

---