6/8/24, 10:46 AM Final Quiz 1: Attempt review

Started on Saturday, 8 June 2024, 10:41 AM

State Finished
Completed on Saturday, 8 June 2024, 10:46 AM
Time taken 4 mins 55 secs
Marks 20.00/20.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

How does the Quickprop algorithm adjust the learning rate for each weight in the neural network?

a. It uses a fixed learning rate for some weights and a variable learning rate for others
b. It uses a fixed learning rate for all weights
c. It adjusts the learning rate based on the previous weight update 
d. It uses a variable learning rate for all weights

Question 2

Correct

Mark 1.00 out of 1.00

What is a deterministic finite state machine?

a. A finite state machine that has multiple possible next states for each combination of current state and input
b. A finite state machine that has multiple possible current states for each combination of next state and input
c. A finite state machine that has a unique current state for each combination of next state and input
d. A finite state machine that has a unique next state for each combination of current state and input 

Question 3
Correct

Mark 1.00 out of 1.00

What are the components of a Markov decision process (MDP)?

a. States, actions, and costs

b. States, outcomes, and costs
c. States, actions, and rewards
d. States, outcomes, and rewards 

https://trimestral.amaesonline.com/2333/mod/quiz/review.php?attempt=118251&cmid=64473 1/6
6/8/24, 10:46 AM Final Quiz 1: Attempt review

Question 4

Correct

Mark 1.00 out of 1.00

What is a finite state machine?

a. A state machine that can be in an infinite number of states

b. A state machine that can only be in a finite number of states 
c. A state machine that can change its state an infinite number of times
d. A state machine that can only be in one of two states

Question 5
Correct

Mark 1.00 out of 1.00

How does the backpropagation algorithm work?

a. By adjusting the activation function of the neural network

b. By adjusting the weights and biases of the neural network in the forward pass
c. By adjusting the weights and biases of the neural network in both the forward and backward passes
d. By adjusting the weights and biases of the neural network in the backward pass 

Question 6

Correct

Mark 1.00 out of 1.00

What are the biases in the backpropagation algorithm?

a. The values that are adjusted in the forward pass

b. The values that are used to determine the activation function of the neural network 
c. The values that are adjusted in the backward pass
d. The values that are used to determine the output of the neural network

Question 7
Correct

Mark 1.00 out of 1.00

How does the RProp algorithm adjust the learning rate?

a. It increases the learning rate if the error increases and decreases the learning rate if the error decreases
b. It uses a predetermined set of learning rates for each iteration
c. It uses a fixed learning rate regardless of the error
d. It increases the learning rate if the error decreases and decreases the learning rate if the error increases 

https://trimestral.amaesonline.com/2333/mod/quiz/review.php?attempt=118251&cmid=64473 2/6
6/8/24, 10:46 AM Final Quiz 1: Attempt review

Question 8

Correct

Mark 1.00 out of 1.00

How does the Quickprop algorithm improve upon traditional gradient descent algorithms?

a. It uses a variable learning rate 

b. It uses a fixed learning rate
c. It uses a smaller learning rate
d. It uses a larger learning rate

Question 9
Correct

Mark 1.00 out of 1.00

What is a decision tree?

a. A neural network used in deep learning

b. A tree-like model used to make predictions in supervised learning 
c. A linear model used to make predictions in supervised learning
d. A clustering algorithm used in unsupervised learning

Question 10

Correct

Mark 1.00 out of 1.00

What are the hidden layers of a feedforward neural network called?

a. Hidden layers 
b. None of the above
c. Input layers
d. Output layers

Question 11
Correct

Mark 1.00 out of 1.00

What are the weights in the backpropagation algorithm?

a. The values that are used to determine the output of the neural network 
b. The values that are adjusted in the forward pass
c. The values that are used to determine the activation function of the neural network
d. The values that are adjusted in the backward pass

https://trimestral.amaesonline.com/2333/mod/quiz/review.php?attempt=118251&cmid=64473 3/6
6/8/24, 10:46 AM Final Quiz 1: Attempt review

Question 12

Correct

Mark 1.00 out of 1.00

What is a disadvantage of using a Q-learning algorithm?

a. It can learn from its mistakes and improve over time

b. It cannot handle high-dimensional data
c. It requires a large amount of data
d. It cannot handle continuous state spaces 

Question 13
Correct

Mark 1.00 out of 1.00

How does the RProp algorithm handle local minima in the optimization process?

a. It avoids local minima by using a fixed learning rate

b. It avoids local minima by using a dynamic learning rate 
c. It avoids local minima by using a predetermined set of learning rates
d. It gets stuck in local minima

Question 14

Correct

Mark 1.00 out of 1.00

How does the RProp algorithm handle the initialization of weights in the optimization process?

a. It uses predetermined weights

b. It uses the weights of the best performing iteration
c. It uses random weights 
d. It uses the previous iteration's weights

https://trimestral.amaesonline.com/2333/mod/quiz/review.php?attempt=118251&cmid=64473 4/6
6/8/24, 10:46 AM Final Quiz 1: Attempt review

Question 15

Correct

Mark 1.00 out of 1.00

What is a Markov decision process (MDP)?

a. A mathematical model for decision-making in situations where outcomes are entirely random
b. A mathematical model for decision-making in situations where outcomes are partly random and partly under the control 
of a decision maker
c. A mathematical model for decision-making in situations where outcomes are partly random and partly determined by past
actions
d. A mathematical model for decision-making in situations where outcomes are entirely under the control of a decision maker

Question 16
Correct

Mark 1.00 out of 1.00

What is a Q-learning algorithm?

a. A type of supervised learning algorithm

b. A type of unsupervised learning algorithm
c. A type of deep learning algorithm
d. A reinforcement learning algorithm that uses a Q-table to store the expected rewards for each action in each state 

Question 17
Correct

Mark 1.00 out of 1.00

How does a backpropagation network learn?

a. By adjusting the weights and biases of the network based on the input and output data 
b. By adjusting the architecture of the network based on the input and output data
c. By adjusting the activation function of the network based on the input and output data
d. By adjusting the number of neurons in the network based on the input and output data

https://trimestral.amaesonline.com/2333/mod/quiz/review.php?attempt=118251&cmid=64473 5/6
6/8/24, 10:46 AM Final Quiz 1: Attempt review

Question 18

Correct

Mark 1.00 out of 1.00

What is a backpropagation network?

a. A type of artificial neural network that uses deep learning

b. A type of artificial neural network that uses supervised learning 
c. A type of artificial neural network that uses reinforcement learning
d. A type of artificial neural network that uses unsupervised learning

Question 19
Correct

Mark 1.00 out of 1.00

How is a state transition function typically represented?

a. As a set of differential equations

b. As a graph or network
c. As a function of state and input 
d. As a table or matrix

Question 20

Correct

Mark 1.00 out of 1.00

How does the Quickprop algorithm handle weight updates that are too large?

a. It increases the learning rate

b. It discards the weight updates 
c. It reduces the learning rate
d. It reduces the weight updates

◄ Module 10

Jump to...

Final Lab Quiz 1 ►

https://trimestral.amaesonline.com/2333/mod/quiz/review.php?attempt=118251&cmid=64473 6/6