• Title: Big O

  • Series: Advent of Mathematical Symbols

  • YouTube-Title: Advent of Mathematical Symbols - Part 12 - Big O

  • Bright video: https://youtu.be/vEUnVm2k2-k

  • Dark video: https://youtu.be/puem_0keuoM

  • Quiz: Test your knowledge

  • PDF: Download PDF version of the bright video

  • Dark-PDF: Download PDF version of the dark video

  • Print-PDF: Download printable PDF version

  • Thumbnail (bright): Download PNG

  • Thumbnail (dark): Download PNG

  • Subtitle on GitHub: aoms12_sub_eng.srt

  • Timestamps (n/a)
  • Subtitle in English

    1 00:00:00,700 –> 00:00:04,821 The mathematical symbol of today is the landau symbol big O.

    2 00:00:05,771 –> 00:00:11,496 For this notation you either see a normal capital O or the slightly open O.

    3 00:00:12,357 –> 00:00:17,429 In either case you always find next to it parentheses with a function.

    4 00:00:18,286 –> 00:00:24,900 So there is a function named for example g and you also often find the variable name, for example x.

    5 00:00:26,157 –> 00:00:31,714 Now it is important to note that this big O only makes sense with respect to a limit.

    6 00:00:32,857 –> 00:00:37,476 For example you can write at the end, that the variable x goes to the number “a”.

    7 00:00:38,543 –> 00:00:44,131 Also it’s allowed that this “a” is the symbol infinity or the symbol - infinity.

    8 00:00:45,229 –> 00:00:50,584 Now usually in this big O notation we use an equality sign in a symbolic way.

    9 00:00:51,429 –> 00:00:58,384 So you would write another function f(x) is equal to this big O of g(x).

    10 00:00:59,286 –> 00:01:02,689 We do this to make calculations easier to write down.

    11 00:01:03,571 –> 00:01:07,865 However the formal correct way would be to use an element sign here.

    12 00:01:08,771 –> 00:01:14,450 Having said that of course the most important part is that you know the meaning of this expression here.

    13 00:01:15,229 –> 00:01:23,614 Indeed it simply means that the function f on the left hand side does not grow stronger than the function g on the right hand side

    14 00:01:23,671 –> 00:01:25,629 when x goes to the number “a”.

    15 00:01:26,757 –> 00:01:32,838 Hence usually this is helpful when you are not able to put in “a” into the 2 functions.

    16 00:01:33,686 –> 00:01:38,042 For example we always have this case when “a” is the symbol infinity.

    17 00:01:39,214 –> 00:01:43,871 Now of course this explanation that f does not grow stronger than g

    18 00:01:43,957 –> 00:01:47,648 can be put into a formula, where we use absolute values.

    19 00:01:48,814 –> 00:01:59,659 So you would say the absolute value of f(x) ia always less or equal than a fixed constant M times the absolute value of g(x).

    20 00:02:01,014 –> 00:02:06,444 and indeed this should hold for all points x in the neighbourhood around “a”.

    21 00:02:07,357 –> 00:02:16,207 In the case that “a” is infinity this means that we have a bound such that all points x above this bound satisfy this inequality.

    22 00:02:17,414 –> 00:02:23,396 Now if you don’t like this constant M, we can also rewrite the whole thing with a limit superior.

    23 00:02:24,443 –> 00:02:28,343 So we have lim sup where x goes to the point “a”.

    24 00:02:29,086 –> 00:02:33,203 and then we take the absolute value of f divided by g.

    25 00:02:34,486 –> 00:02:39,566 and now this lim sup should be a finite number so we can write less than infinity.

    26 00:02:40,700 –> 00:02:46,614 So you see it means the same thing, we compare the growth rate of f with the growth rate of g.

    27 00:02:47,471 –> 00:02:52,614 and in a case that f would grow faster, then this whole thing here would go to infinity.

    28 00:02:53,757 –> 00:02:57,835 So maybe it’s even better for the explanation when we look at an example.

    29 00:02:58,800 –> 00:03:03,438 So lets say we have the polynomial x^2 + x + 2.

    30 00:03:04,371 –> 00:03:11,291 Then we can say this is in big O of x^2, if x goes to infinity.

    31 00:03:12,414 –> 00:03:19,123 However now you know we can also say that the same polynomial is in big O of x^3.

    32 00:03:20,086 –> 00:03:24,548 Of course also in the case when we look what happens when x goes to infinity.

    33 00:03:25,600 –> 00:03:30,316 Indeed this is what you can check now. x^3 grows faster than x^2.

    34 00:03:31,386 –> 00:03:35,867 However because we only have an inequality here both things are correct.

    35 00:03:36,786 –> 00:03:39,376 Ok i think that’s good enough for today.

    36 00:03:40,043 –> 00:03:44,940 Now i really hope that you learned something today and that i can see you in the next video.

    37 00:03:45,140 –> 00:03:45,886 Bye!

  • Quiz Content

    Q1: Which of the following functions is $\mathcal{O}(x^2)$?

    A1: One need more information.

    A2: $f(x) = 4 x^2 - x$

    A3: $f(x) = \exp(x)$

    A4: $f(x) = 5 x^3 - 1$

    A5: $f(x) = \sin(x)$

    Q2: Which of the following functions is not in $\mathcal{O}(x^3)$ with $(x \rightarrow \infty)$?

    A1: One needs more information.

    A2: $f(x) = 4 x^2 - x$

    A3: $f(x) = \exp(x)$

    A4: $f(x) = 5 x^3 - 1$

    A5: $f(x) = \sin(x)$

  • Back to overview page