edited by
462 views
0 votes
0 votes
Finding the running time of the following algorithm.

$Procedure$ $A(n)$
   $\textrm{ if (n}<=\textrm{2) then return 1 ;}$
   $else$
   $Return(A(\left \lceil \sqrt{n} \right \rceil))$
edited by

1 Answer

Best answer
2 votes
2 votes

Please refer this https://gateoverflow.in/1243/gate2007-45

$T(n) = T(\sqrt{n}) +1;$

Let $n = 2^{m} $

$T(2^{m}) = T(\sqrt{2^{m}}) +1;$

$T(2^{m}) = T(2^{m/2}) +1;$

Now let $T(2^{m}) = S(m)$

So $S(m) = S(\frac{m}{2} ) +1$

It is binary search recurrence relation

so $S(m) = O(log m)= O(log log n)$   ($\because n = 2^{m}$)

since $T(n) = T(2^{m}) = S(m)$

T(n) is $O(log log n)$

selected by

Related questions

631
views
1 answers
1 votes
Sajal Mallick asked Nov 28, 2023
631 views
Consider the problem that given a set Sof n integers and another integer x, whether or not there exist two elements in S whose sum is exactly x. What is the worst ... in dynamic programming? Then complexity should be O(n^2).How O(n logn)?
224
views
0 answers
0 votes
244
views
0 answers
0 votes
Sajal Mallick asked Nov 27, 2023
244 views
As we have to select maximal set of “non overlapping” activities. So like job scheduling algo of greedy we can solve it. So according to that complexity must be O(n logn). But ans is (b). Anyone please explain.
1.1k
views
1 answers
1 votes
shikharV asked Nov 15, 2015
1,083 views
The answer to the above problem is A but I am expecting it to be D as constant amount of work is required to solve each subproblem.