CSE 551: Quiz 4 Solutions
Jamison Weber
April 6, 2020
1 Problem 1
Solve the following recurrence relation using any method. Provide your
answer in big-O notation: T(n) = 2T(
n
2
) + log(n) for n > 1, 0 otherwise
...
CSE 551: Quiz 4 Solutions
Jamison Weber
April 6, 2020
1 Problem 1
Solve the following recurrence relation using any method. Provide your
answer in big-O notation: T(n) = 2T(
n
2
) + log(n) for n > 1, 0 otherwise
• T(n) = O(n)
• T(n) = O(nlogn)
• T(n) = O(n
2
)
• T(n) = O(logn)
1.1 Rationale
This recurrence relation can be tricky to solve using iterative substitution
or tree-based methods. It’s best to use the Master theorem here. Recall the
form:
T(n) = aT(n/b) + f(n)
Since f(n) = log(n) = n
where < log2(2) = 1, the Master method gives:
T(n) = O(n
log2(2)) = O(n)
2 Problem 2
Suppose we have a modified version of Merge-Sort that at each recursive level
splits the array into two parts each of size 1
4
and 3
4
respectively. Also, assume
the size of any given input array is a power of four. Give the asymptotic
time complexity of this Merge-Sort variant.
1
• O(nlogn)
• O(n)
• O(logn)
• O(n
2
)
2.1 Rationale
This gives the recurrence relation T(n) = T(
3
4
) + T(
1
4
) + cn. Using the
Akra-Bazzi method: We must satisfy
X
k
i=1
aib
P
i = 1
Where ai
is the coefficent in front of recursive call i and bi
is the divisor of
n in recursive call i. Thus, we have:
1(3
4
) + 1(1
4
)
P = 1 ⇒ P = 1
Now we use the Akra-Bazzi formula to find:
T(n) ∈ Θ(n
P
(1 + Z n
1
g(u)
u
P +1 du)) = Θ(n(1 + Z n
1
cn
n2
dn))
= Θ(n(1 + c
Z n
1
1
n
dn)) = Θ(n + ncln(n)) = Θ(nlog(n))
3 Problem 3
Suppose we modify the combine step of the closest pair of points algorithm
such that distance δ from dividing line L is updated immediately to δ
0 whenever the distance between two points on either side of L is discovered to be
less than δ. In this sense, we allow the two-dimensional range about L
wherein we compare points to assume multiple areas (getting smaller as δ
becomes updated) during the same combine step for each recursive call. Determine whether the following statement is true or false and explain your
reasoning: The time complexity of the closest pair of points algorithm is
guaranteed to be improved by a constant factor.
• False: If δ is reduced only after the last pair of points sorted
by y-coordinate within δ of L is compared, then no additional
benefit will be gained.
2
• False: The number of comparisons made between points within δ of L
would necessarily be the same as without the modification.
• True: If δ is reduced every time a new closest pair of points is found
during the combine step, then it will always be the case that a fewer
number of future comparisons will be necessary after δ is adjusted toδ
0
.
• True: If δ is reduced every time a new closest pair of points is found
during the combine step, then it will sometimes be the case that a fewer
number of future comparisons will be necessary after δ is adjusted to
δ
0
.
[Show More]