Uncertainty:Bayesian networks

Question 1

Conditional Independence

Answer 1

Two variables are conditionally independent if they are independent given the knowledge of a third variable.

Question 2

Bayesian Network

Answer 2

A graphical model representing probabilistic relationships among variables using nodes and directed edges.

Question 3

Directed Acyclic Graph (DAG)

Answer 3

A graph with directed edges and no cycles

Question 4

Conditional Probability Table (CPT)

Answer 4

A table that defines the probability of a node given the values of its parent nodes in a Bayesian network.

Question 5

Causality

Answer 5

One variable’s direct effect on another variable.

Question 6

Correlation

Answer 6

The statistical degree to which two variables change together.

Question 7

Hidden Variables

Answer 7

Variables that are not observed but affect the system in question.

Question 8

Evidences

Answer 8

Known values of observed data used in inference.

Question 9

Joint Distribution

Answer 9

The probability of all possible combinations of values for all variables in a system.

Question 10

Exact Inference

Answer 10

Computing the exact posterior probabilities or distributions over variables given evidence.

Question 11

Enumeration

Answer 11

An exact inference technique that sums over all values of hidden variables.

Question 12

Caching

Answer 12

Storing the results of expensive computations to avoid repeating them.

Question 13

Approximate Inference

Answer 13

A technique of estimating the probability values or distributions of variables.

Question 14

Direct Sampling

Answer 14

An approximate inference method that estimates probabilities based on the frequency of sampled outcomes.

Question 15

Joint Probability Distribution Factorization

Answer 15

Making the joint probability of the network simpler by expressing it as a product of conditional probabilities.