USN AM722I1A
Seventh Semester B.E. Degree Examination
ADVANCED AI and ML
(MODEL QUESTION PAPER)
Time: 3 hrs. Max. Marks: 100
Note: Answer any FIVE full questions, choosing ONE full question from each module.
Q. No. Questions Marks BL/CO
Module I
1 Discuss how rationality of an agent depends on its performance measure,
percept sequence, environment, and prior knowledge for the following
a. i. Satellite Image analysis system 10 CL3/CO1
ii. Medical Diagnosis system.
iii. Self-Driving Car
b. Explain with a neat diagram Utility-based agents and Model based Agent. 10 CL2/CO1
OR
2 Apply the Minimax algorithm for a two-level Tic-Tac-Toe game tree.
a. Draw the game tree, label MAX and MIN nodes, calculate utilities, and 10 CL3/CO1
determine the optimal move.
Formulate the 8-puzzle problem as a state-space search problem. Represent
b. possible moves and describe how a problem-solving agent can reach the goal 10 CL2/CO1
state.
Module II
3 A disease affects 1% of a population. The diagnostic test has 95% accuracy for
both positive and negative results. Using Bayes’ theorem, compute the
a. 10 CL3/CO2
probability that a person has the disease given that they test positive. Show all
steps.
Explain conditional independence and show how it simplifies computation in
b. Bayesian Networks. Construct a simple Weather Prediction Model with 10 CL3/CO2
variables like Cloudy, Rain, Sprinkler, WetGrass and show one inference.
OR
a. In a Wumpus World, the agent perceives a breeze in square (1,2).
4
Using probabilistic reasoning, infer the likelihood of a pit being in (2,2) or 10 CL3/CO2
(1,3). Show reasoning steps and probability estimation.
Write an algorithm for inference using Full Joint Probability Distribution with
b. variables Toothache, Cavity, Catch. Demonstrate how the marginal probability 10 CL3/CO2
of Cavity can be obtained.
Module III
Derive the Gradient Descent update rule for Linear Regression.
5 a. 10 CL3/CO3
Using data points (1, 1), (2,2) learning rate 0.1 and initial weight w=0 perform
Page 1 of 3
Seventh Semester B.E. Degree Examination
ADVANCED AI and ML
(MODEL QUESTION PAPER)
Time: 3 hrs. Max. Marks: 100
Note: Answer any FIVE full questions, choosing ONE full question from each module.
Q. No. Questions Marks BL/CO
Module I
1 Discuss how rationality of an agent depends on its performance measure,
percept sequence, environment, and prior knowledge for the following
a. i. Satellite Image analysis system 10 CL3/CO1
ii. Medical Diagnosis system.
iii. Self-Driving Car
b. Explain with a neat diagram Utility-based agents and Model based Agent. 10 CL2/CO1
OR
2 Apply the Minimax algorithm for a two-level Tic-Tac-Toe game tree.
a. Draw the game tree, label MAX and MIN nodes, calculate utilities, and 10 CL3/CO1
determine the optimal move.
Formulate the 8-puzzle problem as a state-space search problem. Represent
b. possible moves and describe how a problem-solving agent can reach the goal 10 CL2/CO1
state.
Module II
3 A disease affects 1% of a population. The diagnostic test has 95% accuracy for
both positive and negative results. Using Bayes’ theorem, compute the
a. 10 CL3/CO2
probability that a person has the disease given that they test positive. Show all
steps.
Explain conditional independence and show how it simplifies computation in
b. Bayesian Networks. Construct a simple Weather Prediction Model with 10 CL3/CO2
variables like Cloudy, Rain, Sprinkler, WetGrass and show one inference.
OR
a. In a Wumpus World, the agent perceives a breeze in square (1,2).
4
Using probabilistic reasoning, infer the likelihood of a pit being in (2,2) or 10 CL3/CO2
(1,3). Show reasoning steps and probability estimation.
Write an algorithm for inference using Full Joint Probability Distribution with
b. variables Toothache, Cavity, Catch. Demonstrate how the marginal probability 10 CL3/CO2
of Cavity can be obtained.
Module III
Derive the Gradient Descent update rule for Linear Regression.
5 a. 10 CL3/CO3
Using data points (1, 1), (2,2) learning rate 0.1 and initial weight w=0 perform
Page 1 of 3
