0
$\begingroup$

$$T(n) = 2\cdot \sqrt{n} \cdot T(\sqrt{n}) + \Theta (\lg n)$$

I have been trying to solve this question but I could not find anything.

My approach:

$n = 2^k$

$S(k) = T(2^n)$ and $S(k/2) = T(2^{n/2})$

Finally: $S(k) = 2^{1+k/2} \cdot S(k/2) + c \cdot \lg(k) $

After that, I tried to build recursion tree but I can not find the sum. Do you have any ideas?

Thanks in advance.

$\endgroup$
1
  • $\begingroup$ This is not radical enough. You need $n=a^{2^k}$ and correspondingly $S(k)=T(a^{2^k})$. Note that there were already answers given to the identical question stackoverflow.com/q/22101848/3088138. $\endgroup$ Commented Mar 1, 2014 at 17:55

1 Answer 1

0
$\begingroup$

Rewriting $T(n)=2⋅\sqrt{n}⋅T(\sqrt{n})+Θ(\lg n)$ in terms of

$$n=a^{2^k}\ \text{ and }\ S(k)=T(a^{2^k})⋅a^{-2^k}⋅2^{-k}$$

for some $a>1$ results in the recursion

$$S(k)⋅a^{2^k}⋅2^k=2⋅a^{2^{k-1}}⋅S(k-1)⋅a^{2^{k-1}}⋅2^{k-1}+Θ(2^k⋅\lg a)$$

or after collection

$$S(k)=S(k-1)+a^{-2^k}⋅Θ(\lg a),$$

so that

$$S(k)=S(0)+\sum_{j=1}^k a^{-2^j}⋅Θ(\lg a)$$

and this is bounded by the geometric series, and probably sharper upper bounds exist, the estimates of the Newton-Kantorivich theorem look similar.

So essentially, this $S(k)$ is a constant, and $k=\log_2(\log_a(n))$, rendering

$$T(n)=n⋅\log_a(n)⋅S(\log_2(\log_a(n)))=Θ(n⋅\log_a(n))$$

$\endgroup$

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.