Unit1: Introduction to AI:
1. How turing Test is used to evaluate intelligence of a machine? What properties
a machine should have to pass the Total turing test?
2. What is Artificial Intelligence (AI)? Describe your own criteria for computer
program to be considered intelligent.
3. Define with suitable supporting statements and examples, “Artificial
Intelligence is the system that act like humans”.
4. Do you agree “the development of Artificial Intelligence has had some
negative effect on the society”? If you agree list some of them and put your
opinion in the support of development of Artificial Intelligence.
5. How the dimensions like thinking humanly and thinking rationally are used
to evaluate intelligence behavior of a machine.
6. How can you define AI from the dimension of rationality?
7. Justify that “System that think rationally” and “System that act rationally” are
the part of artificial intelligence. Explain it with practical examples.
8. “System that think like humans” and “System that act like humans” are the
part of artificial intelligence. Justify that statement with practical examples.
Unit2: Agents and Environment
1. What do you mean by rational agents? Are the rational agents intelligent?
Explain.
2. For each of the following agents, determine what type of agent architecture is
most appropriate (i.e. table lookup, simple reflex, goal-based or utility based).
a. Medical diagnosis system
b. Satellite image analysis system
c. Part-pricking robot
d. Refinery controller
3. What are intelligent agents? Differentiate Model Based Agents differ from
utility Based Agents differ from utility Based Agent. Mention suitable
examples of each.
4. What is intelligent agent? Design PEAS framework for,
- Soccer playing agent
- Internet shopping assistant
5. Discuss the types of environment where an agent can work on.
6. How agent can be configured using PEAS framework? Illustrate with
example.
Unit3: Informed and Uninformed Search
1. Define backward chaining. Explain the importance of backward chaining with
two practical examples.
2. Construct a state space with appropriate heuristics and local costs. Show that
Greedy Best First search is not complete for the state space. Also illustrate A*
is complete and guarantees solution for the same state space.
3. What do you mean by forward chaining? Why it is required? Explain it with
two practical examples.
4. How informed search are different than uniformed? Given following state
space, illustrate how depth limited search and iterative deepening search
works? Use your own assumption for depth limit.
Hence, S is start and K is goal
5. Explain the uninformed search techniques with example.
6. What is state space representation of problem? Represent the root finding
problem having four cities in to state representation (you can choose any
ordering of cities and links) and devise the complete problem formulation.
7. In problem solving, why problem formulation must follow goal formulation?
How state space representation can be used to solve a problem? Support your
answer with an example.
8. Consider the following graph, steps cost is given on the arrow: Assume that
the successors of a state are generated in alphabetical order, and that there is
no repeated state checking. A is the starting node and C is goal node.
9.
a. Of the four algorithms breadth-first, depth-first and iterative-
deepening, which find a solution in this case?
b. Write sequence of node expanding by algorithm if finds solution.
10. In problem solving, what is the concept of state space, state, successor
function, goal test and path cost? Illustrate each with suitable example.
11. Justify the searching is one of the important part of AI. Explain in detail about
depth first search and breadth first search techniques with an example.
12. Consider the search space below, where S is start state and G 1 and G2 are goal
state. The arcs are labelled with step cost. Given the heuristic by H(~) for each
nodes. Now use iterative depending and greedy best first search for finding
the goal state, Also determine which goal state is reached first in each case.
13. How uniform cost search works? Given following state-space, use uniform
cost
search algorithm to find the goal. Show each of iterations.
i.
Here S is start state and G is goal state.
14. What is heuristic information? Suppose that we run a greedy search algorithm
with h(n) = – g(n) and h(n) = g(n). What sort of search will the greedy search
follow in each case?
15. The minimax algorithm returns the best move for MAX under the assumption
that MIN play optimally. What happens when MIN plays sub optimally?
16. What is meant by admissible heuristic? What improvement is done in A*
search than greedy Search? Prove that A* search gives us optimal solution if
the heuristic function is admissible.
17. How iterative depending search is better than DFS and BFS. Consider
following state space, use iterative deepening search considering S as start and
g as goal.
18. Searching is an important part of AI, justify it. Explain any two types of blind
search with suitable examples. How can you expand it to informed search?
19. Justify that searching is one of the important part of AI. Explain in detail about
depth first search and breadth first search techniques with an example.
20. What is the need of alphabeta pruning in game search? Given following search
space with utility, perform mini-max search and identify alpha-beta cutoff if
any. Play from perspective of max player first.
i.
ii.
21. How searching is done in adverserial search? Given following search space
with utility values perform minimax search for max player and identify the
possible alpha/beta cutoff.
22. Justify that AI can’t exist without searching. Explain in detail about any two
types of informed search with practical examples.
23. Illustrate with an example, how uniform cost search algorithm can be used for
finding goal in a state space.
24. Consider a following state space representing a game. Use minimax search to
find solution and perform alpha-beta pruning, if exists.
25. Given following search space, determine if these exists any alpha and beta
cutoffs.
Unit4: Knowledge Representation:
1. What is Bayes’s theorem? Explain its applications.
2. Consider following facts:
Every traffic chases some driver. Every driver who horns is smart. No traffic
catches any smart driver. Any traffic who chases some driver but does not
catch him is frustated.
3. Now configure FOPL knowledge base for above statements. Use resolution
algorithm to draw a conclusion that "If all drivers horn, then all traffics are
frustated".
4. Why disjunctive normal form is required? Explain all the steps with examples.
5. How resolution algorithm is used in FOPL to infer conclusion?
Consider the facts;
6. Convert following statement into FOPL, every friend of Ramesh has visited
pokhara. Everyone who visits Pokhara does boating on Fewalake. Ramesh has
done boating on Fewalake. Now using resolution try to infer; some friend of
Ramesh has done boating on Fewalake.
7. Why normal forms are required in AI? How do you convert to the disjunctive
normal form? Explain all the steps with practical examples.
8. How do you convert to conjunctive normal form? Explain all the steps with
examples.
9. Differentiate between inference and reasoning. Why probabilistic reasoning
is important in the AI? Explain with an example.
10. State whether the following sentences are valid, unsatisfiable, or neither.
a. Smoke => Smoke
b. Smoke => Fire
c. (Smoke => Fire) => (~Smoke=>~Fire)
d. Smoke V Fire V ~Fire
11. Translate the following sentence into first order logic:
i. “Everyone’s DNA is unique and is derived from their parents’
DNA”.
ii. “No dog bites a child of its owner”.
iii. “Every gardener likes the sun”.
iv. “All purple mushrooms are poisonous”.
12. Briefly describe the approaches of knowledge representation with example.
13. Consider the following statements.
All cats like fish, cats eat everything they like, and Ziggy is a cat.
a) Translate the sentences into FOL.
b) Convert the sentences into clausal normal form.
c) Answer using FOL, if Ziggy eats fish?
14. What is script? How knowledge is represented in script? Illustrate component
of script with a example.
15. Differentiate between inference and reasoning. Why probabilities reasoning
is important in AI? Explain with an example.
16. Consider the following sentence:
[(food => party) V (drinks => party)] => [(food ^ drinks) => party]
a. Convert the right hand and left hand sides of main implication into
CNF.
b. Prove the validity of sentence using resolution.
17. Represent the following sentences into a semantic network.
a. Birds are animals.
b. Birds have feathers, fly and lay eggs.
c. Albatross is a bird.
d. Donald is a bird.
e. Tracy is an albatross.
18. Consider the knowledge base:
“If it is hot and humid, then it is raining. If it is humid, then it is hot. It is
humid”
a. Describe a set of propositional letters which can be used to represent
the knowledge base.
b. Translate the KB into propositional letters using your propositional
letters from part a.
c. Is it raining? Answer this question by using logical inference rule with
KB.
19. Show a possible sequence of two recognize-art cycles. Which will be the new
content of the short-term memory after these two cycles?
20. Define knowledge representation system. How knowledge is represented
using semantic networks? Illustrate with an example.
21. What do you mean by causal network? Explain it with practical application.
22. How Knowledge is represented using scripts? Support your answer with
suitable example.
23. Convert the following sentence into predicate logic.
a. “No dog bites a child of its owner”?
b. “No two adjacent countries have the same color”?
24. Construct sematic network for following facts:
Ram is a person. Person are humans. All humans have nose. Humans are instarces
of mammals. Ram has weight of 60 kg. Weight of Ram is less than weight of Sita.
25. Why disjunctive normal form is required? Explain all the steps with examples.
26. Consider the following statements:
Rabin likes only easy courses. Science courses are hard. All courses in the BIM
are easy. MGT 101 is a BIM course.
a. Translate the sentences into predicate logic.
b. Convert your sentences into clausal normal form (CNF).
27. Define frame. How knowledge is encoded in a frame? Justify with an
example.
28. What are conceptual graphs? Represent the following statements into
conceptual graph.
29. What is the difference between symbolic and non-symbolic AI? Represent the
following knowledge in semantic network.
a. Robin is bird
b. Clyde is a Robin
c. Clyde owns a nest from spring 2014 to fall 2014
30. Represent the following sentences into a semantic network.
a. Birds are animals.
b. Birds have feathers, fly and lay eggs.
c. Albatros is a bird.
d. Donald is a bird.
e. Tracy is an albatross.
31. Convert following statements to FoPL.
a. No teachers are ignorant.
b. Some teachers who are ignorant are not skillful.
c. All skillfull teachers are likely by all.
32. Convert following sentences to FOPL.
If every helper is busy then there is a job in the queue.
A job is in queue but the helper is not busy.
Every helpers are teased by someone.
33. How uncertain knowledge is represented? Given following full joint
probability distribution representing probabilities of having different sizes of
CD, find the probability that a CD cover has a length of 130mm given the
width is 15mm.
Unit5: Learning
1. Describe mathematical model of neural network. What does it means to train
a neural network? Write algorithm for perception learning.
2. Define mathematical model of artificial neural network. Discuss how Hebbian
learning algorithm can be used to train a neural network. Support your answer
with an example.
3. Define learning. Why learning frame work is required? Explain about learning
frame work with block diagram and examples.
4. What does it means to train a neural network? Consider following neural
network, how back-propagation algorithm can be used to train it?
5. Define Learning. Why learning
framework is required? Explain
about learning frame with block
diagram and examples.
6. Why do we require learning? Explain about learning framework with suitable
block diagram and examples.
7. What is learning by induction? Explain inductive learning process with
example.
8. What is machine learning? How genetic algorithm can be used to train agents?
Discuss the operations of genetic algorithm.
9. Configure a feed-forward neural network with your own assumptions of
inputs and weights and express it mathematically. Write an algorithm for
training neural networks using allebbian learning.
10. How concepts of most specific consistent hypothesis and most general
consistent hypothesis are used in learning through examples. How
generalization specialization tree is maintained for these concepts?
11. What is a Neural Network? Explain any one type of neural network with
practical example.
12. Derive the mathematical model of neural network. Explain any one type of
neural network with its algorithm.
13. What is neural network? Explain the neural net learning methods.
14. What is back propagation? Explain all the steps involved in the back
propagation with an example.
15. What do you mean by machine vision? Discuss the components of a machine
vision system.
16. What is crossover operation in genetic algorithm? Given following
chromosomes show the result of one-point and two point crossover.
a. C1 = 01100010
b. C2 = 10101100
17. Choose appropriate crossover point as per your assumption.
18. What is machine learning? Explain the learning from analogy and instance
based learning?
19. Write an algorithm for learning by Genetic Approach.
Unit6: Application of AI
( Note: most of the question has appeared in unit 5 learning chapter)
1. How can you construct expert system? Explain knowledge engineering with
a block diagram.
2. Define a natural language processing. Explain the different issues involved in
the natural language processing.
3. What is an expert system? Explain the architecture and feature of rule-based
expert system.
4. What is natural language processing? How morphological analysis is done
during processing?
5. What do you mean by natural language processing? What is the importance
of pragmatic analysis in NLP?
6. What is expert system? How it works? Mention role of inference engine in
expert system.
7. How can you construct expert system? Explain knowledge engineering with
a block diagram.
8. Knowledge consists of facts, beliefs, and heuristics, justify it. Explain the
advantages and disadvantages of an expert system.
9. Why do we require expert system structure? Draw the block diagram and
explain it with practical example.
10. Differentiate between natural language understanding (NLU) and natural
language generating (NLG). Why we have to study natural language
processing? Explain it.
11. How semantic and pragmatic analysis is done in natural language processing.
12. Define natural language processing. Explain the different issues involved in
the natural language processing.
13. Explain the steps of Natural Language Processing.
14. Explain the different steps involved in the natural language processing (NLP)
with block diagram and examples.
15. Differentiate between natural language understanding (NLU) and natural
language generation (NLG).
16. How the concept of machine vision are used in Robotics to configure sensors
of Robots?
17. How syntactic and semantic analysis is done during natural language
processing? Explain with example.
