Study Exam

BBM402: Study Problems
1. Write the recursive relation for the running time of the algorithm given to solve ’Tower
of Hanoi’ problem and provide a solution for it.
Solution:
T (n) = 2T (n − 1) + 1.
The solution is T (n) = 2n − 1. This can be shown by using induction on n. The proof
by induction should consist of two steps: 1) base case, 2) inductive case, typically
making use of the information called inductive hypothesis. The detailed solution is
in the reading for ’Solving Recurrences’.
2. The running time for the mergesort algorithm is given as
T (n) = T (dn/2e) + T (bn/2c) + O(n).
Use a recursion tree, to show that T (n) = O(n log n). Show all your work.
Solution: (See class notes or the slides or the reading on ’Solving recurrences’ for
the solution.)
3. Give an example to a divide-and-conquer algorithm (recursive) to multiply two n-digit

numbers with O(n2 ) running time. Do you know any other algorithm that runs faster,
meaning with o(n2 ) running time? If yes, describe that algorithm and its running time
as well.
Solution: (See reading on ’Recursion’, Section 1.8)
4. Consider the recurrence relation T (n) = 2T (n/2) + n/ log n. Use a recursion tree, to
show that T (n) = Θ(n log log n).
Solution: See the slides.
1
√
5. Consider the recurrence relation T (n) = T ( n) + 1. Use a recursion tree to show
√ √
that T (n) = Θ(log log n). Similarly, if T (n) = nT ( n) + n, show that T (n) =
Θ(n log log n).
Solution: See the slides and reading on ’Solving Recurrences’.
6. • Karatsuba’s multiplication algorithm has running time T (n), where T (n) = 3T (n/2)+
n. Use a recursion tree to show that T (n) = Θ(nlog2 3 ).
• Let T (n) = 2T (n/2) + n/ log n. Show that T (n) = Θ(n log log n).
• Let T (n) = 4T (n/2) + n/ log n. Show that T (n) = Θ(n2 ).
Solution: See the reading on ’Solving Recurrences’.
7. Q1: Argue briefly why the recursive relation for the running time of the Tower of Hanoi
problem is T(n) = 2T(n-1)+1 for n at least 1, in an recursive algorithm provided in class.
In other words, explain why we cannot do better with a shorter running time or why not
more running time is needed in the given recursive algorithm.
Solution: See reading on recursion and class notes.
8. a) Write a short recursive program to calculate Fibonacci number n.

b) In the recursion tree for this algorithm, give the number of leaves in this tree and
prove that it is correct by induction.
Solution: See reading on dynamic programming.
9. In the conflict-free class-scheduling problem, a) prove that at least one maximal conflict-
free schedule includes the class that finishes first.
b) prove that the greedy algorithm that picks the non-conflicting class that finishes
first from the available classes outputs an optimal schedule (with maximal number of
classes).
2
Solution: See reading on greedy algorithms.
10. Design a Turing machine by writing the steps of the algorithm (no diagram asked) exe-
cuted to accept if the input is in the language A and reject otherwise.
• A = {w#w|w ∈ {0, 1}∗ }

n
• A = {02 |n ≥ 0}
• A = {0n 1n |n > 0}
• A = {0n 1n 2n |n > 0}
Solution: All of the questions are examples in Sipser’s book (see webpage) Chapter
3 and our slides. Therefore, you can find the solutions there.
11. Besides these, I may give you a Turing machine with its diagram and an input, where
you are supposed to list all configurations showing the executions of the Turing machine
and the final state (accept or reject). (See Sipser’s book, exercises 3.1 and 3.2)
12. Show that if the language L and its complement L̄ are recognizable, then L is decidable.
(Lecture Notes)
13. Show that LD is not recursively enumerable (recognizable). (Lecture Notes)
14. Show that AT M is not a decidable language. (pg.174 in Sipser’s book/ Lecture Notes)
15. What is the ’Halting problem’? (pg.174 in Sipser’s book)
16. Describe the (universal) TM that recognizes AT M . (Lecture Notes)
17. Let L be an undecidable language. Is it possible that both L and L̄ (the complement of L)
are recognizable. If yes, give an example. If no, give a short proof. (pg.181 in Sipser’s
book)
18. What is the maximum independent set problem? What is the decision problem equiva-
lent to this problem? (Lecture 7)
19. Show that the maximum independent set problem is in NP. (Lecture 7)
20. What is the minimum vertex cover problem? What is the decision problem equivalent to
this problem? (Lecture 7)
3
21. Show that the minimum vertex cover problem is in NP. (Lecture 7)
22. Show that the independent set problem can be reduced to the clique problem in polyno-
mial time? How big is this polynomial in the worst-case? (Lecture 7)
23. Show that the independent set problem can be reduced to the vertex cover problem in
polynomial time? How big is this polynomial in the worst-case? (Lecture 7)

Study Exam

Uploaded by

Copyright:

Available Formats

Study Exam

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Study Exam

Uploaded by

Copyright:

Available Formats

BBM402: Study Problems

2. The running time for the mergesort algorithm is given as

T (n) = T (dn/2e) + T (bn/2c) + O(n).

3. Give an example to a divide-and-conquer algorithm (recursive) to multiply two n-digit

Solution: (See reading on ’Recursion’, Section 1.8)

Solution: See the slides.

Solution: See the slides and reading on ’Solving Recurrences’.

Solution: See the reading on ’Solving Recurrences’.

Solution: See reading on recursion and class notes.

8. a) Write a short recursive program to calculate Fibonacci number n.

Solution: See reading on dynamic programming.

• A = {w#w|w ∈ {0, 1}∗ }

13. Show that LD is not recursively enumerable (recognizable). (Lecture Notes)

15. What is the ’Halting problem’? (pg.174 in Sipser’s book)

16. Describe the (universal) TM that recognizes AT M . (Lecture Notes)

You might also like