retagged by
5,184 views

1 Answer

Best answer
7 votes
7 votes

We have Stirling's approximation which says

$$\log n! = \Theta( n \log n)$$

So, $(\log n)! = e^{\log (\log n)!} \\=  \Theta \left(e^{ ( \log n \log \log n )}\right) \\=  \Theta \left( \left({e^{\log n}} \right) ^ {\log \log n}\right) \\= \Theta \left(n^{\log \log n}\right)$.

Hence, polynomially lower bounded but not upper bounded. For polynomial, we need $n^c$, where $c$ is a constant, which is not the case here.

$(\log \log n)! = e^{\log (\log \log n)!} \\=  \Theta \left(e^{ ( \log \log n \log \log \log n )}\right) \\=  \Theta \left(\left( {e^{\log \log n}} \right)^ {\log \log \log n} \right) \\= \Theta \left(\left(\log n\right)^{ \log \log \log n}\right) = O(n) \\ \left( \because n^{\log \log n} = O(e^{n}), \\ \text{ detailed at end, and replace }n \text{ with } \log n \right)  $.

Hence polynomially upper bounded.


$n^{\log \log n} = {e^{\log n}}^{\log \log n} \\= e^ {\log n \log \log n}  \\= O(e^ n)$

selected by

Related questions

0 votes
0 votes
0 answers
2
Akriti sood asked Dec 27, 2016
371 views
in an effort to make MERGE-SORT faster, you decide to divide the array into k equal sized, disjoint subarrays, where k 2. This means that you have to merge k lists. How ...
0 votes
0 votes
2 answers
3
0 votes
0 votes
1 answer
4
deepti asked Jul 31, 2016
492 views
It is given in CLRS book chapter 3, page no. 56 that lg^k(n) = (lg n)^k. Can somebody please give me an example or check if my example is correct? Shoudn't lg^k(n) be { l...