Pascal's Triangle - Numberphile

Поділитися
Вставка

КОМЕНТАРІ • 1 тис.

  • @ImAllInNow
    @ImAllInNow 7 років тому +247

    Want another pascal's triangle pattern? Take the top right diagonal (1, 1, 1, ...) and treat it as if it was a decimal number with the decimal point between the first two numbers (1.11111...). Now square that number. You get 1.234567... which equals the second diagonal. Cube it and you get 1.3717421... Which is the third diagonal (you have to smush the numbers together the same was as 1,5,10,10,5,1 = 161,051).
    Now this works in bases other than 10. If you treat 1.11111 as a binary number it equals 2 (same way 1.9999... = 2 in decimal). Now 2^2 = 4. So we know that 1.234567... in binary equals 4. Written as an infinite sequence, 1.23456 in binary is 1 + 2/2 + 3/4 + 4/8 + 5/16 + 6/32 + ... which indeed equals 4!
    This pattern even has a connection to another Numberphile video, Grafting Numbers, and it's truly remarkable, although this comment section is too small to contain it.

    • @geri30
      @geri30 4 роки тому +17

      Lol, fermat’s last theorem reference. I like it.

    • @michaelhoefler5118
      @michaelhoefler5118 4 роки тому +7

      Jeo okay fermat

    • @force10guy26
      @force10guy26 3 роки тому +3

      It'd probably all fit now a days lol

  • @austynhughes134
    @austynhughes134 7 років тому +601

    I love how you guys constantly interview people who have a genuine love of mathematics! It makes these videos so amazing.

    • @jmp01a24
      @jmp01a24 5 років тому +5

      yes they just go around Europe asking random people in the streets.

    • @jedizombiekiller9065
      @jedizombiekiller9065 3 роки тому +6

      I will never forget klein bottle man

    • @zaneroy1255
      @zaneroy1255 3 роки тому

      instablaster

    • @Triantalex
      @Triantalex Рік тому

      ??.

    • @3arezu
      @3arezu 9 місяців тому

      same!!! love this channel sm

  • @ExplosiveBrohoof
    @ExplosiveBrohoof 7 років тому +191

    For anyone interested in why these patterns emerge:
    *Pascal's Triangle encodes n choose k:* This arises from the fact that _C(n, k) = C(n-1, k-1) + C(n-1, k)._ This can be solved algebraically (note that _C(n, k) = n!/(k!(n-k)!)),_ but there's a combinatorial argument to it as well. Say you have _n_ different ice cream flavors and you want to choose _k_ of them for your super tall ice cream cone. Then you can consider two distinct cases: the combinations with chocolate, and the combinations without chocolate. If you include chocolate, then you have _n-1_ flavors left and you need to choose the remaining _k-1_ flavors. If you don't include chocolate, then you still have _n-1_ flavors, but you still have _k_ flavors to choose from as well. Hence, when you add them together, you should get the total _C(n, k)_ combinations. Because of this identity, you can inductively show that Pascal's Triangle encodes _n_ choose _k._
    *The rows are consecutive powers of 2:* Remember that if you want to choose _k_ objects from _n_ items, you go down to the _n_ -th row in the triangle and you go over by _k._ This means that the _n_ -th row will be the numbers _C(n, 0), C(n, 1), C(n, 2), ..., C(n, n)._ Now, consider the total number of ways to choose _n_ objects, regardless of the number of items you choose. This will be the sum of all of the cases where you choose _i_ objects for _0

    • @Maharani1991
      @Maharani1991 7 років тому

      +

    • @GABRIELFILMSTUDIOS
      @GABRIELFILMSTUDIOS 7 років тому +3

      Arbitrary Renaissance Thanks! I'm going to try to come up with an explanation myself for the ones I did not know before actually reading your comment though.

    • @ExplosiveBrohoof
      @ExplosiveBrohoof 7 років тому +4

      When I say it takes a bit of complex induction to work out the Fibonacci pattern, I just mean that it's tough to rigorously describe. The intuition is really easy like you said: the first diagonal grabs all the left sums and the second diagonal grabs all the right sums.
      Regarding Fermat, that's really interesting! Thanks for taking the time to share your discoveries.

    • @breathless792
      @breathless792 7 років тому +3

      I found a simpler way to describe it:
      Consider a string of ones and zeros as both a Pascals triangle row (in mod 2) and as a binary number.
      Given that take the following string
      1000…0001 (between the 1’s on either side are all zeros)
      Each 1 creates its own mini triangle due to the zeros on either side (this partly explains Siepinski’s triangle) at a certain point the two sub triangles will meet, they will meet when the both rows are all the 1’s (i.e. like 1111….111)
      Proof:
      If you consider a row of all 1’s, 1+1=0 (mod 2) so all will be zeros in the next row apart from extra 1’s placed at either side I.e it creates 1000…0001 so the previous row to 1000…0001 will be all 1’s with 1 less of them.
      Now the row with the 1’s can be called row X (which means it has X terms) now if you take each sub triangle to that row (the 1000…0001 is row 1 of the mini triangle) so at this point you will at row 2X so 2X terms so when you put the two row X’s together there are no 0’s in between)
      Now that you’ve done 2X rows you have all 1’s again so it creates another 1000…0001
      Now as binary numbers each row can be factorised as the sub triangle row (as a binary number) multiplied by the 1000…0001 and that represents multiplying that number by all previous rows
      Finally to prove that the numbers you multiply each past term are Fermat numbers
      Proof:
      First the number is in the form 1000…0001 (in binary) which means it’s 1 more than a power of 2 (which in binary is 1000….000)
      Second you take the “111…111” number and duplicate it so the number of 1’s is doubled and if you add 1 it becomes a power of 2 with the same number as zeros as the number that had 1 added had 1’s, add 1 again and it has the same number of digits, and adding the first 1 increased it digits by 1 so if the 111…111 was row X then 1000…0001 (the number obtained by adding 2) it the next row of the Triangle and the number in the middle(not the triangle the value since there is difference is 2) is a power of 2 with a number of 0’s as explained above
      Since you started with 1 on its own (at the top of the triangle) and each time you doubled it
      (1,11,1111…etc) it means that the number of 0’s in the power of two (and therefore the power) is a power of 2 so it is in the form 2^(2^N) since this number has one added to create the number you multiply by, its (2^(2^N))+1 which is the definition of a Fermat number

    • @hyh2012
      @hyh2012 6 років тому +2

      Another way to prove the Hockey Stick Theorem:
      The corresponding identity is:
      C(n,n) + C(n+1,n) + ... + C(n+k,n) = C(n+k+1,n+1) (n,k∈N) (The other direction is just symmetry of Pascal's triangle C(n,r)=C(n,n-r))
      The identity can be proved using a combinatorics argument:
      Number of ways to choose n + 1 items from a set of n + k + 1 elements: C(n+k+1,n+1)
      But this can be computed in another way by considering cases:
      including element 1 - C(n+k, n)
      excluding element 1, including element 2 - C(n+k-1, n)
      excluding elements 1, 2 including element 3 - C(n+k-2, n)
      ...
      Keep going with this pattern until:
      excluding elements 1, 2, ..., k, including element k + 1 - C(n, n)
      If keep going excluding more than k elements then it is impossible to choose n + 1 from n + k +1 elements. So we have covered all cases here.
      Equating the 2 methods of computing number of ways to choose n + 1 items from a set of n + k + 1 elements gives the identity.

  • @3zehnutters
    @3zehnutters 7 років тому +305

    You can verfy if a number n is prime by looking in the n-th row and checking if every number(beside the 1´s) in that line is 0 modulo n.

    • @AnastasisGrammenos
      @AnastasisGrammenos 7 років тому +8

      really? wow

    • @Fightclub1995
      @Fightclub1995 7 років тому +9

      So n is a prime factor of every number in the nth row (except the ones). Cool pattern.

    • @ZardoDhieldor
      @ZardoDhieldor 7 років тому +10

      Pascal's triangle never stops amazing me.

    • @3zehnutters
      @3zehnutters 7 років тому +3

      i was super excited when i first discovered this pattern myself

    • @nuclearcoconut3664
      @nuclearcoconut3664 7 років тому +7

      That makes me wonder...is testing pascal's triangle as efficient as simply testing factors?

  • @AndersHass
    @AndersHass 7 років тому +77

    00:50 "We can go on as long as we want and for me it is when the first double digit shows up because that is when adding gets hard." lol

  • @adamloepker8057
    @adamloepker8057 4 місяці тому +2

    You explained that better than an hour of lecture in 2 minutes, well done!

  • @alonkellner5375
    @alonkellner5375 7 років тому +7

    The nth row in the Pascal triangle discribes the geometrical properties of an n dimensional simplex, for instance, the 3D simplex is a tetrahedron, a tetrahedron has 4 vertices, 6 edges, and 4 faces, which is the third row.
    (**to get the full row count 1 zero dimensional shape, or an 'empty' shape, and 1 3D cell, that gets you to 1,4,6,4,1)

  • @austinbryan6759
    @austinbryan6759 5 років тому +12

    I think it's pretty cool how, with Pascal mod 2, you can see how adding odds and evens work.
    0 + 0 = 0
    1 + 0 = 1
    1 + 1 = 0
    Which maps to the classic rules of adding evens and odds.
    What's more remarkable is that it also follows adding in single digits in binary. 1 + 1 = 10 in binary but if you make the results only the ones place then it's 0

  • @jenecomprends
    @jenecomprends 7 років тому +466

    Please ramble on more about this triangle!

    • @justinward3679
      @justinward3679 7 років тому +27

      Did you know that if you put a point at every edge of Pascal's triangle you get a triangle?

    • @PhilBagels
      @PhilBagels 7 років тому +25

      The first diagonal is all 1's. The second diagonal is the natural numbers. The third diagonal is the triangular numbers. The fourth diagonal is the tetrahedral numbers. The fifth diagonal is the pentachoronal numbers. etc. The nth diagonal is the n-1-dimensional simplex numbers.

    • @ZardoDhieldor
      @ZardoDhieldor 7 років тому +16

      Except you don't, because it's infinite and only has one corner and two edges.

    • @danielketterer9683
      @danielketterer9683 7 років тому

      jenecomprends exactly. more like please don't

    • @TheWeepingCorpse
      @TheWeepingCorpse 7 років тому +2

      Zardo Dhieldor an edge if defined by two vectors, so maybe its a vector and two rays?

  • @khayanjumbe1462
    @khayanjumbe1462 3 роки тому +2

    I used to watch you guys when I was like 6 or 7 and now I can finally understand what you are talking about

  • @ravneetsingh1499
    @ravneetsingh1499 7 років тому +762

    Fibonacci numbers in Pascal triangle..... JUST AWESOME

    • @TheWolf5575
      @TheWolf5575 7 років тому +29

      My triangle is amazing huh?

    • @justinward3679
      @justinward3679 7 років тому +4

      You didn't discover it you thief!

    • @TheWolf5575
      @TheWolf5575 7 років тому +9

      Justin Ward but it has my name :c

    • @aadfg0
      @aadfg0 7 років тому +7

      You stole the triangle from Chinese mathematicians. Thief!

    • @QualityContentX
      @QualityContentX 7 років тому +1

      yee

  • @imveryangryitsnotbutter
    @imveryangryitsnotbutter 7 років тому +34

    A more illustrative way to write out the part at 4:47 would be:
    100,000
    050,000
    010,000
    001,000
    000,050
    000,001
    -------------
    161,051

  • @thanhgoal8711
    @thanhgoal8711 7 років тому +46

    I love it when mathematicians can't hide the hype while talking about numbers!!!!
    All of this is adorable!!!! XD

  • @Maharani1991
    @Maharani1991 7 років тому +4

    One of my favourite Numberphile videos in a long time. I knew Pascal's triangle, but I had no idea it showcased this many math phenomenons. Sierpiński's triangle is what blew my mind the most.

  • @Xeroxias
    @Xeroxias 7 років тому +26

    So cool to see an undergrad on numberphile

  • @yaseen157
    @yaseen157 7 років тому +1

    We've been learning about it's application in chemistry as part of my A-level course - In a high resolution Nuclear Magnetic Resonance imaging spectrum, the number of peaks and their sizes have relative areas described exactly by Pascal's triangle - a peak with 4 splits in it suggests 3 Hydrogen atoms are adjacent to the unique environment of hydrogen observed, and each peak has relative area 1:3:3:1

  • @NeemeVaino
    @NeemeVaino 7 років тому +6

    Reminds a childhood discovery -
    To calculate 5th row for example, without adding up previous ones, just multiply 1 * 5/1 * 4/2 * 3/3 * 2/4 * 1/5.
    Note the numerator decreasing while denominator increases.
    So generally, k-th number in the n-th row is n!/(k!(n-k)!)

    • @yosefmacgruber1920
      @yosefmacgruber1920 5 років тому

      We know that the 1st number in the 5th row is 1. So let's try and see if that is what your formula produces. 5!/(1!(4!)) = 5. Oops. 5 ≠ 1. Are you considering the 1 to be in the k=0 position and the 5 to be the k=1 position? It also appears that you are considering the top of the triangle to be of row n=0?

    • @angelmendez-rivera351
      @angelmendez-rivera351 5 років тому +1

      Yosef MacGruber Yes. That is actually the definition of the Pascal triangle. The topmost row is the 0th row and the leftmost diagonal is the 0th diagonal. The adding recursion is not part of the definition, that's just an easy construction that works due to the properties that choose(n, k) as a function satisfies.

    • @angelmendez-rivera351
      @angelmendez-rivera351 5 років тому

      Yosef MacGruber Therefore, your counterargument is incorrect.

  • @IceMetalPunk
    @IceMetalPunk 7 років тому +113

    New conjecture: Pascal's triangle encodes literally everything that can be encoded.

    • @angelmendez-rivera351
      @angelmendez-rivera351 5 років тому +9

      IceMetalPunk Sure it can. Since Pascal's triangle is of order Aleph(0), the set of things it encodes is of the size of the set of the real numbers, Aleph(1). So at the very least, it must encode almost everything.

    • @samuelking1624
      @samuelking1624 3 роки тому +14

      IceMetalPunk's conjecture: "Pascal's triangle encodes literally everything that can be encoded." would be very funny to see in a paper. You should try to proove.

    • @ingenuity23
      @ingenuity23 2 роки тому +3

      @@samuelking1624 a hundred years later, its a millenium prize problem, and we all were here to witness history right before our eyes

  • @Eltro920
    @Eltro920 7 років тому +9

    Finally, a video about Pascal's triangle, and it's amazing.

  • @RichardDominguezTheMagicIsReal
    @RichardDominguezTheMagicIsReal 7 років тому +1

    honestly what I really like it how excited she is to explain the triangle, her enthusiasm had me interested

  • @NickKravitz
    @NickKravitz 7 років тому +21

    Around the 8th grade in US we learn Pascal's triangle to solve n choose k problems. In high school we learn the short cut formula for each entry: n choose k equals n! / (k!(n-k!)). In university we learned binomial formula containing the n choose k term. This is by far the most practical use for Pascal's triangle.

    • @nathanisbored
      @nathanisbored 7 років тому +6

      i did not learn pascal's triangle in 8th grade in california, or ever in school for that matter. i also never learned the choose function in school, definitely not high school. i did learn binomial theorem in calculus 2 later on tho

    • @tupactheory3739
      @tupactheory3739 7 років тому

      I first was introduced to Pascal's Triangle in the third grade... when I was eight.

    • @anthonyderose6443
      @anthonyderose6443 7 років тому

      Nick Kravitz I learned Pascal's triangle and binomial theorem as a junior in HS to expand expressions. I always chose to use pascals in my work. Never understood binomial theorem.

  • @VernePhilleas
    @VernePhilleas 5 років тому +2

    I love Casandra Monroe's zeal for mathematics. This video was inspiring and enriching. I love Pascal's Triangle and all the mathematical intricacies it reveals!

  • @clover7359
    @clover7359 7 років тому +3

    If you go down the diagonals, you get n dimensional triangle numbers.
    For example, 0 dimensional triangle numbers are 1, 1, 1, 1, ... etc
    1 dimensional triangle numbers would be 1, 2, 3, 4, 5, 6, 7, etc
    2 dimensional triangle numbers would be 1, 3, 6, 10, 15, 21, 28, 36, etc
    Then the 3 dimensional triangle (pyramidal) numbers would be 1, 4, 10, 20, 35, 56, etc and so on. You can keep going in higher dimensions if you just keep going along the next diagonal.

  • @martinepstein9826
    @martinepstein9826 7 років тому +7

    Here's how I heard about the relation between Pascal's triangle and Fibonacci numbers.
    Problem: You're climbing a ladder n rungs high and you always have a choice of climbing 1 or 2 rungs at a time (let's call these actions single steps and double steps). How many ways are there to climb the ladder?
    One approach to the problem is to use combinatorics: first find the number of ways to climb if you never take a double step, then the number of ways if you only take 1 double step, then 2, etc and then add it all up. This sum corresponds to adding terms along a shallow diagonal of Pascal's triangle.
    A second approach is to use recurrence: to climb n rungs you must first climb n-1 rungs and then take a single step OR climb n-2 rungs and then take a double step. So if f(n) is the number of ways to climb n rungs then we have f(n) = f(n-1) + f(n-2) which gives us the Fibonacci sequence.

    • @martinepstein9826
      @martinepstein9826 7 років тому +1

      I remember reading that this problem was first solved by composers of Indian classical music who wanted to know how many ways there are to, say, fill a bar of music with 8th and 16th notes. Not in those western terms of course.

  • @RodeyMcG
    @RodeyMcG 7 років тому +9

    11:09 Brady's little exhale when he gets what's going on is the exact same reaction I made :D Very cool!

  • @sofia.eris.bauhaus
    @sofia.eris.bauhaus 7 років тому +1

    i once made up a "simplex operator": i generalized triangle numbers for any dimension: (triangle (2-simplex), tetrahedron (3-simplex), pentachoron (4-simplex)). the first operand was the edge length and the second operand was the number of dimension.
    x △ 2 was a triangle number for edge length x and so on. as i wrote down the Cayley table (like a multiplication table) i noticed quite a bit of a pattern. suddenly it struck me that i was writing down Pascal's triangle sideways.
    this was one of the coolest things that ocurred to me. :)

  • @UMosNyu
    @UMosNyu 7 років тому +8

    6:26 is my favorite thing. All the mods. Had a great maths teacher who showed this pattern to us.

  • @gdsu234
    @gdsu234 7 років тому +9

    I absolutely love this channel. Everytime the poorly thought out school curriculum kills my interest in math, one of these videos fires it up again. Thanks guys!

  • @enlongchiou
    @enlongchiou 7 років тому +56

    One more interesting thing for Pascal's triangle of (x-1)^n, change sign from 3rd term on at every row turn out is trivial zero of zeta function at s=1, -2 = 1 -2 - 1, -4 = 1 - 3 -3 +1, -6 = 1 - 4 - 6 + 4 - 1, -8 = 1 - 5 - 10 + 10 - 5 + 1,etc...-2n, .nontrivial zero which are extension of trivial zero obey same rule have 2^n series, (2*n)!/(n!)^2 of moment of nontrivial zero 1,2,6,20.. right at middle line of triangle, let bottom of triangle from 0 to 1 at x-axis, 1,2,6,20.. right at x = 1/2 line as Riemann hypothesis predicted.(Euler product of (p-1)/p is 0.04875 from 2 to 99991, take 2^9632 - 1 of mod(10^10,po)/po get 34490000, 0.04875*10^10/34490000 =14.13 po is all possible combination from 2 to 99991, 2nd 0 ,21.02 = 487500000/23190000 get without 2., 487500000/19500000 = 25 3rd zero of zeta function without 2 and 3, so on...25*(1/2)(2/3)(4/5) + 1/2 - 1/6 - 5/10 + 25/30 + 1/3 - 10/15 +0/5.+3 -1 = 9 prime number counting until 25.)

    • @anon8857
      @anon8857 7 років тому +3

      hey enlong chiou! i remember you from g+ ! ur awesome!!

    • @camilogallardo4338
      @camilogallardo4338 7 років тому +5

      enlong chiou give us the nontrivial zeros!

    • @Israel2.3.2
      @Israel2.3.2 7 років тому +10

      A favorite of mine. If this is confusing just google 'Faulhaber's formula.'
      Let x/(exp(x) - 1) = a + (b/1!)x + (c/2!)x^2 + (d/3!)x^3 + (e/4!)x^4 + (f/5!)x^5 + ... [and let mx^n = (m)*(x^n)]
      Then
      1^0 + 2^0 + 3^0 + ... + n^0 = (1/1)(an)
      1^1 + 2^1 + 3^1 + ... + n^1 = (1/2)(an^2 - 2bn)
      1^2 + 2^2 + 3^2 + ... + n^2 = (1/3)(an^3 - 3bn^2 + 3cn)
      1^3 + 2^3 + 3^3 + ... + n^3 = (1/4)(an^4 - 4bn^3 + 6cn^2 - 4dn)
      1^4 + 2^4 + 3^4 + ... + n^4 = (1/5)(an^5 - 5bn^4 + 10cn^3 - 10dn^2 + 5en)
      1^5 + 2^5 + 3^5 + ... + n^5 = (1/6)(an^6 - 6bn^5 + 15cn^4 - 20dn^3 + 15en^2 - 6fn)
      etc.
      Notice the 'truncated' version of Pascal's triangle.

    • @iustinianconstantinescu5498
      @iustinianconstantinescu5498 7 років тому +1

      enlong chiou Amazing!!!

    • @slipknnnot
      @slipknnnot 7 років тому

      Says the anonymous guy called "Fish Bones"

  • @briannalove749
    @briannalove749 7 років тому

    I first learnt about Pascal's Triangle because of how you can use it in the expansion of brackets with a high power. So if you had (x+a)^5, you would go to Row 5 of the triangle, and each value is a coefficient, in order, so you add that to each product, and list the powers in descending order for x and ascending order for a. Will definitely save a lot of time in exams. In January, I had a summer school for my math this year (to prepare us for senior highschool math) and we looked at other ways it can be used then as well (mainly combinations and permutations - I don't remember which). It's such an amazing mathematical tool.

  • @Zeezjay
    @Zeezjay 7 років тому +163

    what if you extended the triangle to a tetrahedron or any simplex of n dimensions?

    • @Sejiko
      @Sejiko 7 років тому +22

      wow this is mind blowing stuff and i have some ideas.... thank you man.

    • @connorp3030
      @connorp3030 7 років тому +30

      I've heard they've done that for a trinomial distribution pattern, pretty neato

    • @Sejiko
      @Sejiko 7 років тому +7

      Imagine a programm that represent this pattern as tetrahedron in a 3d world and highlight some numbers.

    • @abrasivepaste
      @abrasivepaste 7 років тому +27

      Look up Pascal's Pyramid

    • @wellme3367
      @wellme3367 7 років тому +9

      You get the multinomial coefficients.

  • @torlachrush
    @torlachrush Рік тому +1

    Casandra Monroe's enthusiasm is inspiring.

  • @pythagorasaurusrex9853
    @pythagorasaurusrex9853 7 років тому +3

    Great stuff! I knew most of the properties from the beginning of the video. But I NEVER heard about the fact that Fibonacci's row is encoded in the Pascal triangle. Mind blowing!!

  • @2bratsmom
    @2bratsmom 4 роки тому

    I love Pascal's Triangle and all the possibilities hidden in it. I'm from India and I am proud to say that it was originally found in the works of Pingala, the Indian Mathematician, thousands of years ago and his student brought it to Arabia and it travelled West where Pascal made it popular. This is called Meru Prasthara in Sanskrit and it means "Ladder to Mt Meru". In Vedic literature, Mt Meru is so tall that the Himalaya is like a small stone on it. :) Indeed, this triangle is infinite!

  • @pyotrleflegin7255
    @pyotrleflegin7255 7 років тому +7

    Thank you -- your enthusiasm is infectious!

  • @Holobrine
    @Holobrine 7 років тому

    I love the geometric implications in Pascal's triangle. The first row is a point. The second row is a line segment. The 3rd row is a square with a diagonal line (1 point, 2 adjacent points, 1 point). 4th row is a cube with opposite triangular vertex figures (1 point, 3 adjacent points, 3 more adjacent points, 1 point). Beyond that it's hard to explain but it still works and it's BEAUTIFUL!!!
    What's more, these geometric figures can be used as simple diagrams to understand why Pascal's triangle works for binomial expansion. The third row, for example, is an expansion of (a+b)^2. So pick a corner and label it a^2. The two adjacent corners are ab, and the last corner is b^2.

  • @aosteklov
    @aosteklov 7 років тому +213

    she is so cool

  • @samgilson7476
    @samgilson7476 7 років тому +1

    My friend and I discovered the Serpenski triangle that's hidden inside in a bit of a different way. Instead of applying mod2, we found the absolute value of the difference of the two numbers above. In the same way that you add to produce Pascal's Triangle, we subtracted to create this Difference Triangle, and it turned out to be identical to the mod2 serpenski fractal thing.

  • @VirtualMarmalade
    @VirtualMarmalade 7 років тому +25

    This is like magic. I didn't realize Pascal's Triangle had so much in it!

    • @TheWolf5575
      @TheWolf5575 7 років тому

      Its magic :D

    • @gdash6925
      @gdash6925 5 років тому

      This is made by 1st grader additioning

  • @maikopskoy
    @maikopskoy 7 років тому +1

    Wow I dont regret watching this. well done, I'm quite suprised to find this out. Thank you

  • @llamawaffles5559
    @llamawaffles5559 7 років тому +6

    Alright, i have a question if anyone is still reading the comments on this video. I was doing a problem for my Intro Quantum Mechanics class about spin-1/2 particles, and after doing a bit of math, ended up getting what is essentially a 3d square pyramid of numbers (idk if that's right, its 4 triangles, each making a side), except there is no bottom, it just goes on indefinitely as far as you want to extend it (like pascal's triangle). In fact, my professor noticed that the outer triangle of each side of the pyramid IS Pascal's triangle. Which made me curious if there was some overarching recursion relation (or other relation) to predict future rows/squares of the pyramid. If you look at one of the outside triangles, then remove it and look at the triangle beneath it, and continue doing this, this is what you find:
    Triangle 1: 1, 1 1, 1 2 1, 1 3 3 1, 1 4 6 4 1, 1 5 10 10 5 1, etc (Pascals Triangle)
    Triangle 2: 0, 1 1, 4 0 4, 9 2 2 9, 16 10 0 10 16, 25 27 5 5 27 25, 36 56 28 0 28 56 36, etc (what pattern?)
    Triangle 3: 4, 4 4, 1 12 1, 1 15 15 1, 16 8 40 8 16, 64 0 56 56 0 64, etc (my computer cant compute any more)
    Triangle 4: 0, 9 9, 36 0 36, 64 24 24 64, etc (computer cant do anything layer 10 or below)
    Triangle 5: 36, 36 36, etc
    That is all my computer can do, but as soon as you get to layer 10 of the pyramid, i go beyond the Integer limit in C# and i haven't fixed that problem yet, so my computer just gives me either 64 or null. weird bug, but yeah. if you can figure out some pattern, that would be awesome. And for anyone wondering what these numbers are here is the slightly longer story:
    Each layer, or square, of the pyramid corresponds to the spin of a particle, starting with zero. row zero is a spin-0 particle, row 1 is a spin-1/2 particle, row 2 is a spin-1 particle, row 3 a spin-3/2 particle etc. and the numbers each column/row correspond to the probability that the spin will be measured at that magnitude in a direction orthogonal to the currently known spin (the closes thing Quantum Mechanically to "random"). for example, row 3, column 5, in layer 7, of this pyramid corresponds to the probability that a spin-7/2 particle measured to have spin-5/2 in some direction will be measured to have spin-3/2 in an orthogonal direction. Also the probability that a spin-7/2 particle with spin 3/2 will be measured to have spin-5/2 in an orthogonal direction. However, it is easy to see that the rows do not add to unity, and that is because i have removed a normalization constant to make them all whole numbers. the normalization constant for the n-th layer is simply 2^(-n).
    Any more questions i would be happy to answer.

    • @htmlguy88
      @htmlguy88 7 років тому

      you could download pari gp and try it with high precision. though you may be relating it to pascal's simplex. en.wikipedia.org/wiki/Pascal%27s_simplex

  • @slebeig3035
    @slebeig3035 7 років тому

    It's great to have those kind of videos of brilliant people for free available for everyone, Thank you!

  • @pancakeparliament
    @pancakeparliament 7 років тому +4

    im so happy numberphile finally did a video on pascal's triangle

  • @J7Handle
    @J7Handle 7 років тому +2

    Pascal's triangle describes the coefficients of the resulting polynomial from (a + b)^n
    A 3 dimensional Pascal's pyramid made to represent the coefficients of the resulting polynomial from (a + b + c)^n is related in really cool ways to the standard Pascal's triangle and I would have liked some more details on that.

  • @connorp3030
    @connorp3030 7 років тому +6

    Pacals triangle patterns make more sense if you think of each diaganol as a cumulative frequency of the diagonal before it
    0+0+0+0+0+0+0+0+0...
    0+1 +1 +1 +1 +1 +1...
    0+1 +2 +3+4 +5...
    0+1+3+6+10...
    0+1+4+10...
    0+1+5...
    0+1...
    0...
    One of the patterns I found was that the sum of squares up to n=((nC1)x(n+1C2))-(n+1C3)
    so for example, sum of squares up to 4=(4x10)-10
    They're are patterns for individual square numbers and cube numbers too, but they're way to complicated for me to explain. It makes me wonder if they're are patterns for every exponent, and they're just too complicated to find.
    Also pascals triangle can be used to find the constants terms for any binomial expansion, a pascals pyramid can be used to find the constant terms for a nominal expansion.
    Pascals triangle is real neato.

    • @connorp3030
      @connorp3030 7 років тому +3

      oh, also the diagonals are numbers which are needed to construct a triangle based shape, so the third row is triangular numbers, the fourth row is tetrahedral numbers, and then whatever the fourth dimentional equivalent is is the fourth row I think.

  • @robbes7rh
    @robbes7rh 2 роки тому +1

    Uncanny and completely amazing that what at first blush feels like a children's exercise contains all these other relationships and qualities,

  • @gavinmann4152
    @gavinmann4152 4 роки тому +9

    12:15 'there are still other things i can ramble on about....'
    Me: TELL ME NOWWWW

  • @painovoimaton
    @painovoimaton 4 роки тому

    It is really quite a remarkable thing. So many patterns within it! Pascal's triangle really exemplifies mathematical beauty to me, it has so much going on within a simple rule.

  • @ErnestMSaenz
    @ErnestMSaenz 7 років тому +3

    Here's one of the patterns that I've used to teach multiplication to elementary school students:
    0 9, 1 8,2 7, 3 6, 4 5, 5 4, 6 3, 7 2, 8 1, 9 0 In the first column on the left, write the numbers in ascending order from zero up to nine and in the second column in descending order from nine to zero and, presto!, you've got the nine's multiplication table. I have patterns for the fours, sixes, sevens, eights and also division.

  • @michaelledesma1386
    @michaelledesma1386 7 років тому +1

    Casandra Monroe was amazing in this video!!! Bring her back for more please!!

  • @TakeWalker
    @TakeWalker 7 років тому +9

    She's wearing an Autobot shirt, she is immediately the best guest you've ever had. :D

  • @atrumluminarium
    @atrumluminarium 7 років тому

    For a function f(x)=u(x)v(x), the nth-derivative denoted by f{n} is given by:
    f{n}=∑(nCi u{i}v{n-i}) for i going from 1 to n, and nCi is the binomial coefficient.
    Yes one can generalise the product rule for differentiation to any nth derivative where it turns out that the coefficients are from Pascal's triangle.

  • @estelonb.j.eastham3195
    @estelonb.j.eastham3195 7 років тому +10

    When I saw the Fibonacci Sequence in that I actually said, "It's so beautiful!"

  • @OrangeC7
    @OrangeC7 7 років тому +2

    I think that the reason why there are so many things in pascal's triangle is because it encompasses the one thing almost all of our math is based on, addition. This means that since most of our math has it's roots at addition, we'll find many things inside of the summitive nature of this triangle.

  • @joeytje50
    @joeytje50 7 років тому +11

    9:00 So... the primes' locations on the Pascal triangle are located exactly on the Fibonacci sequence's locations, ignoring 1. They occur at locations 2, then 3, then 5, and the next one (not imaged) would occur at the 8th row. So that's another way to get Fibonacci from Pascal's triangle.

    • @JeSuisUnKikoolol
      @JeSuisUnKikoolol 7 років тому +12

      The 8th row is 255 and is not prime. The next prime (257) is the 9th row

    • @ImAllInNow
      @ImAllInNow 7 років тому +1

      yeah, the primes are at (assuming the top is the 0'th row) locations 1, 2, 4, 8, 16, ...

    • @KafshakTashtak
      @KafshakTashtak 7 років тому

      They are Fermat primes, and so far we only know 5 Fermat primes. Which means the rest of the 2^n rows doesn't create a prime number as long as we know.

  • @rajeevk440
    @rajeevk440 Рік тому +1

    write pascals triangle and use it to calculate powers of 11
    11^3= 1331
    11^4=14641
    11^5= 1 5 (10) (10) 5 1
    start from right last dt =1,
    2nd last =5,
    3rd last but digit is 10 so we take only unit digit =0,
    4th last digit is (10+1 carry from last ) 11 ,so unit dt of 11 =1,
    5th last dt is 5+1 carry from last last digit =6
    6th last digit is same as no carry is done = 1
    so 11^5=161051
    11^6=1 6 (15)(20)(15) 6 1 =1771561
    it works like a basic addition trick (with diffrent rules ) when you know pascals triangle

  • @leobekayombo8087
    @leobekayombo8087 7 років тому +6

    Smart mathematician doing beautiful stuff!!!

  • @harmidis
    @harmidis 7 років тому +1

    Yea Maths is fantastic! You can always be amazed by learning new ways to interpreter Pascal's Triangle!

  • @Sagitarria
    @Sagitarria 7 років тому +9

    it goes waaay back into Hindu, Buddhist, and Jain mathematics as "Mount Meru"
    long before pascal it was known to Pingala in or before the 2nd century BC
    much to explore in it's relationship to cellular automata.

  • @marcjoshuaobidos7555
    @marcjoshuaobidos7555 5 років тому

    It can also be used as the coefficients of terms of the product of the nth power of a binomial, (a+b)^n.
    As of row 2,
    (a+b)^2 = *1* a^2 + *2* ab + *1* b^2,
    and the powers of a decreases in order, and the powers of b increases in order.
    Also works for row 3.
    (a+b)^3 = *1* a^3 + *3* a^2 b + *3* a b^2 + *1* b^3
    Thanx for reading my comment!
    Edit: I like this video!

  • @nachoqt
    @nachoqt 7 років тому +5

    I've never seen the Fibonacci pattern before. Mind blowing.

  • @vijayetc
    @vijayetc 7 років тому

    Extremely interesting. I got curious and searched online for PI and e, and no surprises, they are hidden in this triangle as well. Amazing!. Thanks for this.

  • @pedroscoponi4905
    @pedroscoponi4905 7 років тому +19

    I always thought pascal's triangle was kind of boring. My mistake! hehe XD
    Has anyone ever tried to apply the same methods to different bases?

    • @pedroscoponi4905
      @pedroscoponi4905 7 років тому +2

      I'd love to see what those look like, out of silly curiosity

    • @1987Videolover
      @1987Videolover 7 років тому

      Indicotherium its happened on every base... if u put a number n on the formula, you get (n+1)^x... x means row..
      for examples..
      n = 3, x = 2, u'll get 1(3^2) + 2 (3) + 1 = 9+6+1 = 16 = 4^2
      n = 4, x = 3, u'll get 1(4^3) + 3(4^2) + 3(4) +1 = 64 + 48 + 12 + 1 = 125 = 5^3
      and so on

  • @seanehle8323
    @seanehle8323 7 років тому +2

    I love her passion and energy.

  • @Phalc0n1337
    @Phalc0n1337 7 років тому +3

    Can you show a Parker Triangle next?

  • @dariotavaresdecastroneto1282
    @dariotavaresdecastroneto1282 2 роки тому

    Lovely video 🙂. As a comment, I would say that complete information about the prime factorization of any positive integer n can be obtained from the entries in the n-th row of Pascal's triangle. If frac( x ) represents the fractional part of x, and v_p(n) denotes the exponent of the prime p in the factorization of n, then we can apply the formula:
    v_p(n) = p * Sum_[ j = 1..floor(log_p(n)) ] frac( binomial(n, p^j) * p^(j - 1) / n )
    ref: journal INTEGERS (22) _paper number: 61

  • @TheKrevit
    @TheKrevit 7 років тому +439

    nth

  • @Minecraftster148790
    @Minecraftster148790 7 років тому +2

    If you take a row, put each number in binary and write each number vertically each ending at the same line then the shape of that will be a parabola which gets better and better the further down the triangle you go. There's a nice animation on wikipedia

  • @Lockirby2
    @Lockirby2 7 років тому +3

    It's cool to see an undergraduate on here. :) Gotta love the enthusiasm!

  • @leonardomaranon
    @leonardomaranon 7 років тому +1

    When you do it in mod 2, then the number of 1's in each row is always a power of two. Great video!

  • @jetlag1488
    @jetlag1488 7 років тому +893

    Illuminati confirmed

    • @hiveinsider9122
      @hiveinsider9122 7 років тому +25

      Get back to being a meme on Etho's channel! :P

    • @QualityContentX
      @QualityContentX 7 років тому +1

      We take the Triangle and push it somewhere else

    • @yaseen157
      @yaseen157 7 років тому +13

      lol who'd have known you'd leave Etho's channel to find your way to a top comment here?

    • @H0kram
      @H0kram 7 років тому +6

      Yes. Once again.
      After millions of confirmations.
      This confirms it again.

    • @joeydunn930
      @joeydunn930 7 років тому +10

      Are you following me? I just came from Project Ozone #29... :)

  • @robertvdhill367
    @robertvdhill367 7 років тому +1

    I love so many things about this video!

  • @doctorwho2311
    @doctorwho2311 7 років тому +3

    there is also the binomial thing

  • @lucasdepetris5896
    @lucasdepetris5896 4 роки тому

    Pascal's triangle is amazing! one year ago I used it in order to find tan(n*arctan(x)) in terms of f/g where n is a whole number and f and g are polinomials

  • @lrnzccc
    @lrnzccc 7 років тому +5

    in Italy we call it "Tartaglia's triangle"

  • @gordonfrohman250
    @gordonfrohman250 7 років тому

    Thanks to you i instantly solved a circuits problem involving infinite resistances organized as a pascal triangle. Thanks!

  • @altus1253
    @altus1253 7 років тому +43

    *Claimed by Nintendo LLC

  • @sergiorome48
    @sergiorome48 2 роки тому +1

    I love this, quite useful for my discrete math course

  • @adrianbornabasic7499
    @adrianbornabasic7499 7 років тому +5

    Parker triangle - the new Parker square

  • @Radditz770
    @Radditz770 Рік тому

    Another neat part of the triangle is how it so immediately simplifies the (a+b)^x formula and just gives you the constant factor in front. No need to remember, nor do manual multiplying to remember!
    (a+b)^0 = 1
    (a+b)^1 = 1a+1b
    (a+b)^2 = 1a^2 +2ab +1b^2
    (a+b)^3 = 1a^3 + 3ba^2 + 3ab^2 + 1b^3
    (a+b)^4 = 1a^4 + 4ba^3 + 6(ab)^2 + 4ab^3 + 1b^4
    And so on :D

  • @prateekgurjar1651
    @prateekgurjar1651 7 років тому +627

    pascal newton and Einstein were playing hide and seek. Einstein said to newton "found you!" but Newton went and stood in a square of length one meter and said "Hey I am newton per meter square..you found pascal" HAHAHAHAHAH...ha..ha

    • @thatoneguy9582
      @thatoneguy9582 7 років тому +3

      Prateek Gurjar I don't get it.

    • @prateekgurjar1651
      @prateekgurjar1651 7 років тому +56

      units of pressure is pascal..which is Newton per meter square

    • @sinistrolerta
      @sinistrolerta 7 років тому +9

      That One Guy "Newtons per meter squared" is a unit of pressure which is also known as Pascal

    • @prateekgurjar1651
      @prateekgurjar1651 7 років тому +1

      exactly !

    • @soufian2733
      @soufian2733 7 років тому +17

      I think in english we say "square meter" and not "meter sqaure". I could be wrong

  • @ryanflaws
    @ryanflaws 7 років тому +1

    Some may have already said this but if you use each slot as a single digit number and carry the remainder over dont have to do that huge calculation at 4:45
    I hope that makes sense

  • @mojosbigsticks
    @mojosbigsticks 7 років тому +23

    I'm in love.

    • @mojosbigsticks
      @mojosbigsticks 7 років тому

      A little bit of both.

    • @NoriMori1992
      @NoriMori1992 7 років тому

      Mojos Bigstick A little bit of both what?

    • @mojosbigsticks
      @mojosbigsticks 7 років тому +1

      There was previous reply asking if I was in love with Ms Monroe or with Pascal's triangle. I'm a little bit smitten with both.

    • @denjam2423
      @denjam2423 7 років тому +1

      I was worried you fell in love with the triangle only.

  • @CacchiusMan
    @CacchiusMan 7 років тому

    This video just gave me even more things i did not know about Pascal's triangle. Like, i knew about the diagonal sequences, the powers of two and the factorial coefficient, but i did not knew about the rest. Truly amazing how many things this triangle hides :D Thank you very much brady :)

  • @BerMaster5000
    @BerMaster5000 7 років тому +379

    420 Blaise it

  • @andreygutsuleac2242
    @andreygutsuleac2242 7 років тому +2

    I've been watching Numberphile for quite a while and I really enjoy the videos. But today I discovered I wasn't even subscribed....
    It got me by surprise so I immediately clicked the button.

    • @numberphile
      @numberphile  7 років тому +7

      +Andrew Gutsuleac you you also used the little bell to get notifications!

  • @stevethecatcouch6532
    @stevethecatcouch6532 7 років тому +3

    Cassandra, it wasn't nice to tease us with three or four Fermat primes without mentioning that the run of primes is interrupted after 2^2^4 + 1.

  • @01111011111101etc
    @01111011111101etc 7 років тому

    This is a nice pattern derived from the Pascal's triangle, when one of the edges (1 1 1..) is cut off and alternation between positive/negative values is added, and the numbers are then used as coefficients in polynomials. The resulting polynomials are the gnomons of the perfect powers.
    1
    2 -1
    3 -3 1
    4 -6 4 -1
    etc
    n = sum(k=1..n)(1)
    n² = sum(k=1..n)(2k - 1)
    n³ = sum(k=1..n)(3k² - 3k + 1)
    n⁴ = sum(k=1..n)(4k³ - 6k² + 4k - 1)
    etc

  • @HyperQbeMusic
    @HyperQbeMusic 7 років тому +8

    I wouldn't be surprised if Pi showed up in that triangle in a way… somewhere. 🤔

    • @harinandanrnair6768
      @harinandanrnair6768 7 років тому +3

      HyperQbeMusic u just gave me something to work on... thanks

  • @browniesnofrownies4843
    @browniesnofrownies4843 5 років тому +2

    Excellent explanations!

  • @zevonmxic468
    @zevonmxic468 7 років тому +4

    Mathimus Prime...

  • @kateinmadison
    @kateinmadison 7 років тому

    I've loved Pascal's Triangle since I first learned about it in 7th grade. I had so much fun playing with it in mod 2, mod 3, etc. Different pattern each time!

  • @dQw4w9WgXc
    @dQw4w9WgXc 7 років тому +75

    The Legend of Zelda reference \(°-°)/

  • @WildStar2002
    @WildStar2002 7 років тому

    It's also related to the long-hand method for taking whole-number roots. I don't have time to go into detail, but I stumbled across the relationship when I was trying to figure out how to take third roots by hand. Square roots by hand is related to the 1-2 line as you multiple through and third roots to the 1-3-3 line, and so on, 1-4-6-4 for 4th roots, etc. :-)

  • @MarcoBeri
    @MarcoBeri 7 років тому +64

    The right name is Tartaglia's Triangle :-)

    • @claudiuacsinte4757
      @claudiuacsinte4757 7 років тому +23

      Facciamoci sentire ahaha

    • @matt-vi2pn
      @matt-vi2pn 7 років тому +12

      Marco Beri quindi non sono l'unico italiano qui

    • @ChristianJiang
      @ChristianJiang 7 років тому +10

      Ehi, Triangolo di Yang Hui, ci siamo arrivati prima noi :-)

    • @tgvv2980
      @tgvv2980 7 років тому +4

      stavo pensando l'esatta stessa cosa... perché questi tizi lo stanno chiamando "triangolo di pascal".....?

    • @hdman511
      @hdman511 7 років тому +2

      Volevo proprio scriverlo..

  • @jackren295
    @jackren295 7 років тому

    Finally!! One of my least understood mathematical construct which appears everywhere and is extremely useful is finally here!! Expecting more explanations on its relations with permutation and arrangements. Also, Greeting From China!

    • @jackren295
      @jackren295 7 років тому

      Didn't expect to see Fibonacci sequence AT ALL!

  • @Someone-cr8cj
    @Someone-cr8cj 7 років тому +65

    damnnn Casandra Monroe back at it again with
    a transformers shirt... .rly?

    • @pressplayhomie
      @pressplayhomie 7 років тому

      Someone My exact sentiments.

    • @CastelDawn
      @CastelDawn 7 років тому +53

      Would be appropriate if it was a video about primes.

    • @Someone-cr8cj
      @Someone-cr8cj 7 років тому +1

      yeah not the best shirt for a mathematician to wear... it takes your focus of the actual video

    • @dermathze700
      @dermathze700 7 років тому +9

      I didn't even notice it until the 8 minute mark or so.

    • @Someone-cr8cj
      @Someone-cr8cj 7 років тому

      pure mathematics joke...

  • @Robi2009
    @Robi2009 7 років тому +1

    I gave thumb up for a Autobot logo t-shirt before even watching :)
    I love Pascal's traingle, I remember trying to get more and more rows during boring mathematics classes in high school ;)
    Just one minor nit-pick: you forgot to mention (or show if I missed it) most common way of using the triangle, that is the n-row numbers are coefficients in (x + y)^n formula, i.e. row 2 means (x + y)^2, which is x^2+2xy+y^2

  • @TheTruthSentMe
    @TheTruthSentMe 7 років тому +47

    It's inevitable to find patterns if you look for them.

    • @Sejiko
      @Sejiko 7 років тому +1

      search for pattern is like a hidden background door in maths.you could find something useful for physiks or other topics.

    • @ramiel555
      @ramiel555 7 років тому +57

      especially when they're actually there

    • @Nixitur
      @Nixitur 7 років тому +21

      There is a world of difference between seeing patterns from random noise and being able to _prove_ there's patterns in something like this.

    • @brokenwave6125
      @brokenwave6125 7 років тому +4

      TheTruthSentMe Sure you can choose to select and omit things at your choosing and then find any pattern.
      but when ever single number in a sequence is used, in a systematic way...it's not a made up pattern. it's really encoded in the numbers.
      just like everything in this video.

    • @1st_ProCactus
      @1st_ProCactus 7 років тому

      Sometimes you don't even have to focus on finding a pattern. People find patterns that don't really exist everyday almost.

  • @randomuser778
    @randomuser778 7 років тому +2

    Fantastic episode! Bloody brilliant. Love the presenter's enthusiasm too!