CPTS

Question 1

What does CPT stand for in the context of probabilistic graphical models?

Answer 1

Conditional Probability Table

Question 2

True or False: A Noisy-Or model allows for the representation of independent causes leading to an effect.

Answer 2

True

Question 3

In a Tabular CPT, how are probabilities typically represented?

Answer 3

As a table that lists all possible combinations of parent variables and their corresponding probabilities.

Question 4

Fill in the blank: The Noisy-Or model generalizes the __________ model.

Answer 4

Or

Question 5

What is the primary purpose of a Sigmoid function in CPTs?

Answer 5

To transform a linear combination of inputs into a probability between 0 and 1.

Question 6

Multiple Choice: Which of the following is NOT a characteristic of a Logical/Deterministic CPT?

Answer 6

It outputs probabilities instead of binary outcomes.

Question 7

What is the output range of a Sigmoid function?

Answer 7

0 to 1

Question 8

True or False: In a Noisy-Or model, each cause can independently contribute to the effect.

Answer 8

True

Question 9

Short Answer: What does a Tabular CPT require when there are multiple parent nodes?

Answer 9

It requires a row for each combination of parent node states.

Question 10

Fill in the blank: The Noisy-Or model is useful in situations where the causes are __________.

Answer 10

not mutually exclusive

Question 11

Multiple Choice: Which equation represents the output of a Sigmoid function?

Answer 11

1 / (1 + e^(-x))

Question 12

True or False: A deterministic relationship in a CPT means that the output is uncertain.

Answer 12

False

Question 13

What does the term ‘noisy’ refer to in a Noisy-Or model?

Answer 13

The presence of uncertainty or randomness in the effect caused by the causes.

Question 14

Short Answer: In a Logical CPT, what is the output when all parent nodes are true?

Answer 14

True

Question 15

Fill in the blank: In a Tabular CPT, the probabilities must sum to __________ for each configuration of parent variables.

Answer 15

1

Question 16

Multiple Choice: Which of the following is an advantage of using a Noisy-Or model?

Answer 16

It simplifies the representation of multiple independent causes.

Question 17

True or False: A Sigmoid function can be used to model binary outcomes.

Answer 17

True

Question 18

What is the key feature of a Logical/Deterministic CPT?

Answer 18

It produces a definitive output based on the input states.

Question 19

Short Answer: What is the typical use case for a Tabular CPT?

Answer 19

To represent relationships in small networks with discrete variables.

Question 20

Fill in the blank: The Noisy-Or model assumes that each cause has an associated __________.

Answer 20

probability of failure

Question 21

Multiple Choice: In a Noisy-Or model, if a cause is present, what is the effect’s probability?

Answer 21

It depends on the probabilities of all present causes.

Question 22

True or False: A Sigmoid function is linear in its output.

Answer 22

False

Question 23

What type of data structure is often used to represent a Tabular CPT?

Answer 23

A matrix or table format.

Question 24

Fill in the blank: In Bayesian networks, CPTs are used to define the __________ of each node.

Answer 24

conditional probabilities

Question 25

Multiple Choice: Which of the following best describes the relationship modeled by a Noisy-Or?

Answer 25

A probabilistic relationship among multiple independent causes.

Question 26

True or False: The output of a Sigmoid function can exceed 1.

Answer 26

False