EXACT INFERENCE IN BAYESIAN NETWORKS

This is an exact inference illustration in Bayesian Networks based on the example given in Figure 14.1, pp511, in [1].

% To help my students (Mathematical Department) make sense of Probabilistic Reasoning in my AI class at Xiangtan University, I thereby write down this code segment, hoping to be helpful.

This is an exact inference illustration in Bayesian Networks based on the example given in Figure 14.1, pp511, in [1].

This problem is structured using a Bayesian Network, where each node represents a variable, and edges represent conditional dependencies. The key variables are:

1. Burglary (B) → Whether a burglary has occurred.
2. Earthquake (E) → Whether an earthquake has occurred.
3. Alarm (A) → The alarm goes off, which could be triggered by either a burglary or an earthquake.
4. JohnCalls (J) → John calls if he hears the alarm.
5. MaryCalls (M) → Mary calls if she hears the alarm.

B E

\ /

A

/ \

J M

As mentioned in Chapter 14, the basic task for any probabilistic inference system is to compute the posterior probability distribution for a set of query variables, given some observed event—that is, some assignment of values to a set of evidence variables.

The first edition is finished by Chixin Xiao, on 16 Feb 2025, 02:45 am, in Changsha, China.

Email: [e-mail]

[1] "Artificial Intelligence: A Modern Approach, 3rd US ed." Accessed: Dec. 28, 2023. [Online]. Available: http://aima.cs.berkeley.edu/