diff options
-rw-r--r-- | main.tex | 32 |
1 files changed, 16 insertions, 16 deletions
@@ -461,7 +461,7 @@ And when the limit of two functions $f, g$ converge to some constant $c$, then $ | |||
461 | \section{Big $\mathcal{O}$ and $\Omega$}% | 461 | \section{Big $\mathcal{O}$ and $\Omega$}% |
462 | \label{sec:big_o_and_omega_} | 462 | \label{sec:big_o_and_omega_} |
463 | 463 | ||
464 | % question a ALMOST DONE {{{1 % | 464 | % question a ✅ {{{1 % |
465 | 465 | ||
466 | \begin{question} | 466 | \begin{question} |
467 | Let $f(n)$ and $g(n)$ be asymptotically positive functions. Prove or disprove the following conjectures. | 467 | Let $f(n)$ and $g(n)$ be asymptotically positive functions. Prove or disprove the following conjectures. |
@@ -484,7 +484,7 @@ If we rearrange the right hand side of the Proposition~\ref{eq:a_bigo}; | |||
484 | 0 \le \frac{1}{c} f(n) \le g(n) \\ | 484 | 0 \le \frac{1}{c} f(n) \le g(n) \\ |
485 | \end{equation*} | 485 | \end{equation*} |
486 | 486 | ||
487 | or simply, | 487 | Or simply, |
488 | 488 | ||
489 | \begin{equation} | 489 | \begin{equation} |
490 | \label{eq:a_bio_rearranged} | 490 | \label{eq:a_bio_rearranged} |
@@ -503,21 +503,11 @@ Has to be true, yet from Equation~\ref{eq:a_bio_rearranged} we know that $f(n)$ | |||
503 | $f(n) = \mathcal{O}((f(n))^{2})$ | 503 | $f(n) = \mathcal{O}((f(n))^{2})$ |
504 | \end{info} | 504 | \end{info} |
505 | 505 | ||
506 | % TODO: this is dumb ask this <05-11-20, yigit> % | ||
507 | |||
508 | % https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-046j-introduction-to-algorithms-sma-5503-fall-2005/assignments/ps1sol.pdf | 506 | % https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-046j-introduction-to-algorithms-sma-5503-fall-2005/assignments/ps1sol.pdf |
509 | 507 | ||
510 | The question text states that $f(n)$ is a positive function. | 508 | We can reuse the definition of Big $\mathcal{O}$ notation given in Equation~\ref{eq:a_bigo}, and using the question text, we have; |
511 | |||
512 | By the definition of Big $\mathcal{O}$ notation, we have; | ||
513 | 509 | ||
514 | \begin{equation*}% | 510 | For $c > 1$ and $n_{0} > 1$; |
515 | \exists ~ c > 0 \quad \text{and} \quad n_{0} \ge 0 \mid ~ 0 \le f(n) \le c \cdot g(n) \quad \forall n \ge n_{0} | ||
516 | \end{equation*} | ||
517 | |||
518 | For $c > 1$ and $n_{0} > 1$ we have; | ||
519 | |||
520 | % TODO: this makes zero sense? <06-11-20, yigit> % | ||
521 | 511 | ||
522 | \begin{align*}% | 512 | \begin{align*}% |
523 | f(n) &\le f(n)^{2} \quad \forall n \ge n_{0} \\ | 513 | f(n) &\le f(n)^{2} \quad \forall n \ge n_{0} \\ |
@@ -525,6 +515,8 @@ For $c > 1$ and $n_{0} > 1$ we have; | |||
525 | f(n) &\le c\cdot g(n) \quad \forall n \ge n_{0} | 515 | f(n) &\le c\cdot g(n) \quad \forall n \ge n_{0} |
526 | \end{align*} | 516 | \end{align*} |
527 | 517 | ||
518 | For $f(n) \ge 1$. The conjecture is true for asymptotically positive function $f(n)$ but does not hold for an $f(n) < 1$. | ||
519 | |||
528 | \begin{info}[] | 520 | \begin{info}[] |
529 | $f(n) + o(f(n)) = \Theta(f(n))$ | 521 | $f(n) + o(f(n)) = \Theta(f(n))$ |
530 | \end{info} | 522 | \end{info} |
@@ -556,10 +548,18 @@ It's trivial to pick $c_1 < 1$ to deal with the left hand side of the inequality | |||
556 | f(n) \le f(n) + g(n) | 548 | f(n) \le f(n) + g(n) |
557 | \end{equation*} | 549 | \end{equation*} |
558 | 550 | ||
559 | For the right hand side, we can pick $n_{0} = 1$ and $c = 1$ in the Equation~\ref{eq:a_little_o}; keeping $g(n) < f(n)$. | 551 | For the right hand side; we can start by picking $n_{0} = 1$ and $c = 1$ in the Equation~\ref{eq:a_little_o}; keeping $g(n)$ strictly smaller than $f(n)$. |
552 | |||
553 | In order to prove the right hand side of the inequality; we can pick $c_2 > 2$ which equals; | ||
560 | 554 | ||
561 | Finally, picking $c2 > 2$ proves the conjecture. | 555 | \begin{eqnarray*}% |
556 | \label{eq:c2gt2} | ||
557 | f(n) + g(n) \le 2 \cdot f(n) \\ | ||
558 | \cancel{f(n)} + g(n) \le \cancel{2 \cdot} f(n) \\ | ||
559 | g(n) \le f(n) | ||
560 | \end{eqnarray*} | ||
562 | 561 | ||
562 | Which we proved above using Equation~\ref{eq:a_little_o}. This conjecture is \emph{true}. | ||
563 | 563 | ||
564 | % 1}}} % | 564 | % 1}}} % |
565 | 565 | ||