Given $n$ balls placed in $m$ boxes, prove that if $n < \frac{m(m-1)}{2}$ then at least two boxes have same number of balls in them.
Application of Pigeon-Hole Principle to balls in bins.
777 Views Asked by Bumbble Comm https://math.techqa.club/user/bumbble-comm/detail AtThere are 4 best solutions below
On
The sum of $m$ distinct non-negative integers is $\geq 0 + 1 + 2 + \dots + (m-1)$. One line of proof is to list the integers in increasing order.
On
The statement is equivalent to "if no two bins have the same number of balls, then $n\ge m(m-1)$."
Let $a_0$ be the number of bins with $0$ balls, $a_1$ the number with $1$ ball, and so on. Then $$ n=\sum_{j=0}^\infty ja_j. $$ If no two bins have the same number of balls, then $a_j=0$ or $1$ for every $j$. In that case, exactly $m$ of the numbers $a_j$ equal $1$. Use this to get the lower bound on $n$.
Alternatively, if one wants to apply the pigeonhole principle explicitly, one can do the following. Label the balls with the numbers $1$ through $n$. Label the bins with the numbers $1$ through $m$. One is free to carry out the labeling of bins so that bin $2$ has at least as many balls as bin $1$, bin $3$ has at least as many balls as bin $2$, and so on. Assume this has been done. Let $b_j$ be the occupancy of bin $j$. By assumption $b_1\le b_2\le\ldots\le b_m$. Define $[1,k]$ to be the set $\{1,2,\ldots,k\}$.
Now there are $\binom{m}{2}=\frac{m(m-1)}{2}$ ways to choose two distinct bins. If the occupancies $b_j$ are all distinct, the we can define the map $f:\{(i,j)\in[1,m]^2\mid j>i\}\rightarrow[1,n]$ by $$ \begin{aligned} f((i,j))=&\text{the label of the $(b_i+1)^\text{st}$ ball in bin $j$, with the balls in bin $j$ ordered}\\ &\text{according to their labels,} \end{aligned} $$ and one can see that $f$ is one-to-one. Now apply the pigeonhole principle with the pairs of distinct bins as the pigeons and the ball labels as the pigeonholes.
This is easier without the pigeonhole principle. Consider any arrangement of balls in $m$ boxes such that no two boxes have the same number of balls. Then at most one box can be empty, requiring $0$ balls. At most 1 additional box can have just one ball; the rest ($m-2$ boxes) have at least two balls. At most 1 box can have 2 balls, and so on. So the $m$ boxes must contain at least $0+1+2+\ldots + (m-1) = \frac{m(m-1)}{2}$ balls.