Scribd
Upload
105 views16 pages

Module 5 - AI

Module 5 discusses uncertain knowledge and reasoning, focusing on probability concepts and their applications in scenarios like the Wumpus World. It covers basic probabilistic notation, Bayes' theorem, and the calculation of conditional and posterior probabilities. The module also includes a question bank for further exploration of these concepts in artificial intelligence and real-world situations.

Uploaded by

Hjhgh Hjhhjj
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
105 views16 pages

Module 5 - AI

Module 5 discusses uncertain knowledge and reasoning, focusing on probability concepts and their applications in scenarios like the Wumpus World. It covers basic probabilistic notation, Bayes' theorem, and the calculation of conditional and posterior probabilities. The module also includes a question bank for further exploration of these concepts in artificial intelligence and real-world situations.

Uploaded by

Hjhgh Hjhhjj
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Module 5

Uncertain Knowledge & Reasoning

Acting under Uncertainty:

Example:

CAR

Nature of Uncertain Knowledge:


Probability Notation:
Concept of Probability:
Basic Probabilistic Notation:
1) Proposition

2) Atomic Event
3) Unconditional Probability

4) Independence

5) Conditional Probability:
6) Full Joint Distribution of multiple variables
7) Inference Using Full Joint Distribution:
Axioms of Probability:
Using the Axioms

Proof:
The Probability of sample space is 1:
P(S) = 1
For any event A, the event ‫ ﬢ‬A is its complement.
A and ‫ ﬢ‬A cover the whole sample space:
P(A) + P(‫ ﬢ‬A) = P(S) = 1
It can be re arranged as
P(A) = 1- P(‫ ﬢ‬A)
Baye’s Rule/Theorem:

Baye’s Rule Formula


Using Baye’s Rule (Examples 1):
Example 2:
Example 3:

Wumpus World Revisited:


To calculate the probability that each of the three squares contains a pit in a
Wumpus World scenario, we typically use Bayesian probability. Lets break this
down step by step.
P(W|S) = P(S|W) x P(W)
P(S)
P(S|W) is the probability of sensing stench if a Wumpus is nearby.
P(W) is the prior probability of a Wumpus being in the neighbouring cells.
P(S) is the probability of sensing a stench in general.

Problem Statement
• We are in a Wumpus World with three adjacent squares (A, B, and C).
• Each of these squares has an initial probability of containing a pit, .
• We have sensed a breeze, which indicates that at least one of these
three squares contains a pit.
• Our goal is to calculate the probability that each of these three squares
(A, B, and C) contains a pit given the breeze.
Step 1: Define Initial Probabilities

Step 2: Calculate Conditional Probability

Step 3: Posterior Probability Calculation


• For each square, the probability of it containing a pit given the breeze is
calculated as:

• The same calculation applies to squares B and C because they are


identical in this scenario:

Conclusion
• Given the presence of a breeze, each of the three squares (A, B, and C)
now has a 33.3% probability of containing a pit.
• This approach can be generalized to any number of adjacent squares
with similar conditions.
--------------------------------

Question Bank
1) In a city 30% of the population owns a dog, while 70% owns a cat.
Among dog owners, 80% takes their dogs for daily walk, and among the
cat owners only 50% do so. If a person is observed walking their pet
daily, calculate probability that this person owns a dog. State a Baye’s
theorem.

2) Explain uncertain knowledge in the context of artificial intelligence.


Discuss the challenges an agent can focus when acting under uncertainty
with the example of diagnosing a dental patient’s toothache.

3) Explain the concept of inference using full joint probability in the context
of agents acting under uncertainty with an example of the following
variables:
Weather = {sunny, rain, cloudy, snow}, Cavity = {cavity, ‫ﬢ‬cavity}
Also calculate the following:
P(cavity v toothache), P(cavity|toothache), P(‫ﬢ‬cavity|toothache)
Given the following full joint distribution for the toothache, cavity, catch
world.

4) Prove all kolomogorovs axioms (Probability axioms).

5) Given that bus arriving late = 0.3, and a student oversleeping probability
is 0.4. Find the probability that the students gets late.

6) Explain marginalization and normalization with full joint distribution of


(toothache, catch, cavity).
7) Explain the role of probability in solving Wumpus world.

8) Explain basic probability notation in detail.

9) Explain Bay’s rule and its use in detail.


10)Explain Independence in quantifying uncertainty with example.
---------------------------------

You might also like