Scribd
Upload
2 views

Fall2022

The document outlines the final examination details for the CS 212 Nature of Computation course at Habib University, including instructions for the exam, topics covered, and specific problems to be addressed. It includes definitions and solutions related to computability, the halting problem, mapping and polynomial time reducibility, the P vs. NP problem, and complexity. The exam consists of 6 problems totaling 75 points, with students required to attempt 65 points.

Uploaded by

Breeha Qasim
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

Fall2022

The document outlines the final examination details for the CS 212 Nature of Computation course at Habib University, including instructions for the exam, topics covered, and specific problems to be addressed. It includes definitions and solutions related to computability, the halting problem, mapping and polynomial time reducibility, the P vs. NP problem, and complexity. The exam consists of 6 problems totaling 75 points, with students required to attempt 65 points.

Uploaded by

Breeha Qasim
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Habib University – City Campus

Course: CS 212 Nature of Computation


Instructors: Asma Larik (L1), Waqar Saleem (L2), Abdullah Zafar (L3)
Term : Fall 2022
Examination: Final Term Date: 12 Dec, 2022
Total Marks: 65 Duration: 135 minutes

Instructions:

1. You are allowed to keep only the following items with you on your table: writing utensils,
stationery box, your HU ID card, your “cheat sheet”, and a drinks container. Please deposit
other items, e.g. bags and devices at the front of the classroom.
2. Make sure that your HU ID card is clearly visible on your table.
3. Your cheat sheet must meet the requirements shared earlier. Non-compliant cheat sheets
will be confiscated.
4. Please ensure sufficient writing utensils in working condition. Do not bother others during
the exam.
5. The problems in this exam rely on argumentation. Make sure to provide sound justifications
that are both precise and concise.

6. If your answer to a problem exceeds 10 lines, it is very likely that you are on the wrong
track.
7. Please submit this problem paper and your cheat sheet with your answer book.
8. This exam consists of 6 problems for 75 points, printed on 5 sides.

9. Attempt any 65 points. It is not necessary to attempt all parts of a problem.


10. You may attempt the problems in any order provided that they are clearly labeled.
11. In case you attempt more than the required points, only the first 65 points will be graded.

viel Spaß und viel Glück!


CS 212, Fall 2022 Final Examination 12 Dec, 2022

1. Computability [35 points]


For each of the prompts below, provide short definitions, comments, or (counter-) arguments
as appropriate.
(a) 5 points Turing decidability.

Solution: A language, L, is Turing decidable if there exists a Turing Machine, M ,


that accepts every w ∈ L and rejects every w 6∈ L.

(b) 5 points Turing recognizability

Solution: A language, L, is Turing decidable if there exists a Turing Machine, M ,


that accepts every w ∈ L.

(c) 5 points The relation between computability and Turing recognizability or decidability.

Solution: A decidable language is computable. A Turing machine that decides it


will eventually halt. A recognizable language may not be computable as a Turing
machine that recognizes it may not halt.

(d) 5 points Every language is computable.

Solution: Saying that a language is computable is to say that it is decidable. Not


all languages are decidable. So the statement is False.

(e) 5 points Every problem can be expressed as a language.

Solution: A language that expresses a problem can be checked to contain or not


contain a given string. The answer therefore is binary. Not every problem has a
binary answer. Thus, not every problem can be expressed as a language.

(f) 5 points A Turing machine can perform any possible computation.

Solution: The Church-Turing thesis establishes that any algorithm with a finite
number of steps in which every step does a finite amount of work can be expressed
as a Turing machine. If we consider algorithms that involve an infinite amount of
work, whatever tease may be, they will be beyond the scope of a Turing machine.
We have already encountered examples of languages that cannot be recognized, let
alone decided, by a Turing machine.

(g) 5 points The complement of AT M is Turing recognizable.

Solution: We prove by contradiction that AT M is not Turing recognizable.

Proof. Assume that AT M is Turing recognizable.


It is known that AT M is Turing recognizable.

Page 2 of 5
CS 212, Fall 2022 Final Examination 12 Dec, 2022

Then AT M is decidable.
But AT M is not decidable. ⊥

2. Halting Problem [5 points]


What is the halting problem and what, in your view, is its significance?

Solution: The halting problem is to determine whether a given algorithm will eventually
halt on a given input. It turns out that the answer is no. In general, given an algorithm
and an input, it cannot be determined whether the algorithm will halt. This conveys
that there is at least one problem that a computer cannot solve. And if there is one,
there may be others. This result is also counter-intuitive, as it originally seems that the
problem must have an easy positive answer. Also, once we know that one problem is
undecidable, we can use it along with reducibility to argue about the undecidability of
other problems.

3. Mapping Reducibility [5 points]


What is meant by mapping reducibility and what good is it to know that a language A is
mapping reducible to another language B?

Solution: A language, A, is mapping reducible to another language, B, if there is a


computable function, f : Σ∗ → Σ∗ , where w ∈ A ⇐⇒ f (w) ∈ B. A is said to reduce to
B or be mapping reducible, or simply reducible to B. If that is the case, then if B can
be solved, then so can A. And if A cannot be solved, then neither can B. This can be
used to solve A using known solutions of B, and to prove that B is undecidable if A is
known to be undecidable.

4. Polynomial Time Reducibility [5 points]


What is meant by polynomial time reducibility and what good is it to know that a language
A is polynomial time reducible to another language B?

Solution: A language, A, is polynomial time reducible to another language, B, if there


is a polynomial time computable function, f : Σ∗ → Σ∗ , where w ∈ A ⇐⇒ f (w) ∈ B.
If A is polynomial time reducible to B, then if B can be solved in polynomial time, then
so can A. And if A cannot be solved in polynomial time, then neither can B. This can
be used to find polynomial time solutions to A using known polynomial time solutions
of B, and to prove that B cannot be solved in polynomial time if A is known to be
unsolvable in polynomial time.

5. P vs. NP [10 points]


(a) 5 points What is the P vs. NP problem and what, in your view, is its significance?

Page 3 of 5
CS 212, Fall 2022 Final Examination 12 Dec, 2022

Solution: Problems in P are those that can be decided in polynomial time. Prob-
lems in NP are those that can be verified in polynomial time. P is a subset of
NP. The problems in NP that are also in P can be decided in polynomial time.
It is not known whether the other problems in NP can be decided in polynomial
time or not. The P vs. NP problem is to decisively conclude whether such prob-
lems can or cannot be decided in polynomial time. The best known algorithms to
such problems all take exponential time. A solution to the P vs. NP problem will
either result in polynomial time algorithms for problems that previously required
exponential time, or establish that the search for such algorithms is futile.

(b) 5 points What is NP-completeness and how can it assist in finding a solution to the
P vs. NP problem?

Solution: An NP-complete problem is a problem in NP to which all other NP


problems reduce in polynomial time. To solve the P vs. NP, an argument needs to
be made about every problem in NP which is a daunting task. As all NP problems
reduce to NP-complete problems in polynomial time, the arguments can instead
be made for any NP-complete problem instead.

6. Complexity [15 points]


(a) 10 points Prove that the language, F ACT , is in P where

F ACT = {hx, ni | n! is the closest factorial lesser than or equal to x}

Solution: We prove that F ACT is not in P by showing that any algorithm for it
cannot run in polynomial time.

Proof. In general, the algorithm to decide F ACT needs to compute n!. For this, it
needs the magnitude of n in order to compute the O(n) products for n!. But what
it is given is hni, the string representation of n on the tape. The magnitude of the
number is an exponential function of the length of its representation. (This was
also the insight in WC 10). Therefore, the algorithm is not in P.

(b) 5 points Prove that the language COLOR, is in NP where

COLOR = {G | there is a way of assigning colors to the vertices of the graph G, such
that no two adjacent vertices are assigned the same color, and the entire graph can be
colored with 4 colors}

Solution: The problem can be restated as deciding that a given graph, G, is 4-


colorable. We prove that COLOR is in NP by showing that it can be verified in
polynomial time.
Proof. Let the certificate for the verifier be a coloring, C, of G, i.e. a function from
the vertices in G to the set of colors. The size of C is O(V ) where V is the number

Page 4 of 5
CS 212, Fall 2022 Final Examination 12 Dec, 2022

of vertices in G. Following is a verifier whose complexity is a polynomial function


of V .
colors ← {} # empty set
for (v, cv ) ∈ C do
colors.add(cv )
for (w, cw ) ∈ C where (v, w) is an edge in G do
colors.add(cw )
end for
end for
if len(colors) ≤ 4 then
ACCEPT
else
REJECT
end if

Page 5 of 5

You might also like