Monday, March 29, 2021

2007 IMO Problems And Solutions

 

Problem 01

Real numbers $a_1, a_2, \dots , a_n$ are given. For each $i$ ($1\le i\le n$) define

\[d_i=\max\{a_j:1\le j\le i\}-\min\{a_j:i\le j\le n\}\]

and let

\[d=\max\{d_i:1\le i\le n\}\].

(a) Prove that, for any real numbers $x_1\le x_2\le \cdots\le x_n$,

\[\max\{|x_i-a_i|:1\le i\le n\}\ge \dfrac{d}{2}   (*)\]

(b) Show that there are real numbers $x_1\le x_2\le x_n$ such that equality holds in (*)

Solution

Since $d_i=\max\{a_j:1\le j\le i\}-\min\{a_j:i\le j\le n\}$, all $d_i$ can be expressed as $a_p-a_q$, where $1\le p\le i\le q \le n$.


Thus, $d$ can be expressed as $a_p-a_q$ for some $p$ and $q$$1\le p\le q\le n$

Lemma) $d\ge 0$

Assume for contradiction that $d<0$, then for all $i$$a_i \le \max\{a_j:1\le j\le i\}\le \min\{a_j:i\le j\le n\}\le a_{i+1}$

$a_i\le a_{i+1}$

Then, ${a_i}$ is a non-decreasing function, which means, $\max\{a_j:1\le j\le i\}=a_i$, and $\min\{a_j:i\le j\le n\}\le a_{i+1}=a_i$, which means, ${d_i}={0}$.

Then, $d=0$ and contradiction.

a)

Case 1) $d=0$

If $d=0$$\max\{|x_i-a_i|:1\le i\le n\}$ is the maximum of a set of non-negative number, which must be at least $0$.

Case 2) $d>0$ (We can ignore $d<0$ because of lemma)

Using the fact that $d$ can be expressed as $a_p-a_q$ for some $p$ and $q$$1\le p\le q\le n$$x_p\le x_q$

Assume for contradiction that $\max\{|x_i-a_i|:1\le i\le n\}<\dfrac{d}{2}$.

Then, $\forall x_i$$|x_i-a_i|<\dfrac{d}{2}$.

$|x_p-a_p|<\dfrac{d}{2}$, and $|x_q-a_q|<\dfrac{d}{2}$

Thus, $x_p>a_p-\dfrac{d}{2}$ and $x_q<a_q+\dfrac{d}{2}$.

Subtracting the two inequality, we will obtain:

\[x_p-x_q>a_p-a_q-d=a_p-a_q-a_p+a_q=0\]

$x_p>x_q$ --- contradiction ($p\le q \rightarrow x_p\le x_q$).


Thus, $\max\{|x_i-a_i|:1\le i\le n\}\ge\dfrac{d}{2}$

(b)

A set of ${x_i}$ where the equality in (*) holds is:

\[x_i=\max\{a_j:1\le j\le i\}-\frac{d}{2}\]

Since $\max\{a_j:1\le j\le i\}$ is a non-decreasing function, $x_i$ is non-decreasing.

$\forall x_i$ :

Let $a_m=\max\{a_j:1\le j\le i\}$$a_m-a_i<a_m-\min\{a_j:i\le j\le n\}=d_i$.

Thus, $0\le a_m-a_i \le d$ ($0\le a_m-a_i$ because $a_m$ is the max of a set including $a_i$)


$|x_i-a_i|=\left|a_m-\dfrac{d}{2}-a_i\right|=\left|(a_m-a_i)\dfrac{d}{2}\right|$

$0\le a_m-a_i\le d$

$-\dfrac{d}{2} \le (a_m-a_i)\dfrac{d}{2} \le \dfrac{d}{2}$

$\left|(a_m-a_i)-\dfrac{d}{2}\right|=|x_i-a_i|\le \frac{d}{2}$


Since $\max\{|x_i-a_i|:1\le i\le n\}\ge\dfrac{d}{2}$ and $|x_i-a_i|\le \frac{d}{2}$ $\forall x_i$$\max\{|x_i-a_i|:1\le i\le n\}=\dfrac{d}{2}$

Problem 02

Consider five points $A,B,C,D$, and $E$ such that $ABCD$ is a parallelogram and $BCED$ is a cyclic quadrilateral. Let $\ell$ be a line passing through $A$. Suppose that $\ell$ intersects the interior of the segment $DC$ at $F$ and intersects line $BC$ at $G$. Suppose also that $EF=EG=EC$. Prove that $\ell$ is the bisector of $\angle DAB$.

Solution

[asy] import cse5; import graph; import olympiad; dotfactor = 3; unitsize(1.5inch);  path circle = Circle(origin, 1); draw(circle);  pair D = (-sqrt(3)/2, -0.5), C = (sqrt(3)/2, -0.5); //G = bisectorpoint(C, B, D); pair Ee = rotate(38,C)*D; pair E = IP(C--Ee, circle,1); pair Gg = rotate(76,C)*D; path circle2 = Circle(E, length(C-E)); pair G = IP(C--Gg, circle2, 1); pair F = IP(C--D, circle2, 1); pair Bb = rotate(-104,C)*D; pair B = IP(C--Bb, circle, 1);  pair A = extension((-1,B.y),(1,B.y),G,F); draw(circle2, dashed); draw(A--G); draw(C--D--A--B); draw(G--B);  draw(E--F); draw(E--C); draw(E--G);  dot("$C$", C, dir(30)); dot("$D$", D, SW); dot("$G$", G, SE);  dot("$E$", E, SW); dot("$F$", F, SW); dot("$A$", A, SW); dot("$B$", B, SE);  draw(D--E,dashed); draw(B--E,dashed);  [/asy]

Since $\angle{DAF}=\angle{CGF}$$\angle{BAF}=\angle{CFG}$, it suffices to prove $CF=CG$.

Let $\angle{FCE}=\alpha$$\angle{GCE}=\beta$$\angle{CDE}=\gamma$. We have:\[CF=2CE\cos{\alpha}, CG=2CE\cos{\beta}\]so,\[\dfrac{CF}{CG}=\dfrac{\cos{\alpha}}{\cos{\beta}}\]Meantime, using Law of Sines on $\triangle{DEC}$, we have,\[\dfrac{CE}{\sin{\gamma}}=\dfrac{DC}{\sin{(180-\alpha-\gamma)}}=\dfrac{DC}{\sin{(\alpha+\gamma)}}\]Using Law of Sines on $\triangle{BEG}$, and notice that $\angle{CBE}=\angle{CDE}=\gamma$, we have,\[\dfrac{CE}{\sin{\gamma}}=\dfrac{BG}{\sin{(180-\beta-\gamma)}}=\dfrac{BG}{\sin{(\beta+\gamma)}}\]so,\[\dfrac{DC}{BG}=\dfrac{\sin{(\alpha+\gamma)}}{\sin{(\beta+\gamma)}}\]Since $\triangle{GFC} \sim \triangle{GAB}$, and $DC=AB$, we have, $\dfrac{DC}{BG}=\dfrac{AB}{BG}=\dfrac{CF}{CG}$. Hence,\[\dfrac{\sin{(\alpha+\gamma)}}{\sin{(\beta+\gamma)}}=\dfrac{\cos{\alpha}}{\cos{\beta}}\]or,\[\sin{(\alpha+\gamma)}\cos{\beta}=\sin{(\beta+\gamma)}\cos{\alpha}\]\[\dfrac{1}{2}(\sin{(\alpha+\gamma+\beta)}-\sin{(\alpha+\gamma-\beta)})=\dfrac{1}{2}(\sin{(\alpha+\gamma+\beta)}-\sin{(\beta+\gamma-\alpha)})\]\[\sin{(\alpha+\gamma-\beta)}=\sin{(\beta+\gamma-\alpha)}\]There are two possibilities: (1) $\alpha+\gamma-\beta = \beta+\gamma-\alpha$, or (2) $\alpha+\gamma-\beta = 180 - (\beta+\gamma-\alpha)$. However, (2) would mean $\gamma=90$, then $EC$ would be a diameter, and $EF < EC$ because $F$ is inside the circle, so (2) is not valid. From condition (1), we have $\alpha=\beta$, therefore $CF=CG$

Problem 03

In a mathematical competition some competitors are friends. Friendship is always mutual. Call a group of competitors a clique if each two of them are friends. (In particular, any group of fewer than two competitors is a clique.) The number of members of a clique is called its size. Given that, in this competition, the largest size of a clique is even, prove that the competitors can be arranged in two rooms such that the largest size of a clique contained in one room is the same as the largest size of a clique contained in the other room.

Problem 04

In $\triangle ABC$ the bisector of $\angle{BCA}$ intersects the circumcircle again at $R$, the perpendicular bisector of $BC$ at $P$, and the perpendicular bisector of $AC$ at $Q$. The midpoint of $BC$ is $K$ and the midpoint of $AC$ is $L$. Prove that the triangles $RPK$ and $RQL$ have the same area.

Solution

The area of $\triangle{RQL}$ is given by $\dfrac{1}{2}QL*RQ\sin{\angle{RQL}}$ and the area of $\triangle{RPK}$ is $\dfrac{1}{2}RP*PK\sin{\angle{RPK}}$. Let $\angle{BCA}=C$$\angle{BAC}=A$, and $\angle{ABC}=B$. Now $\angle{KCP}=\angle{QCL}=\dfrac{C}{2}$ and $\angle{PKC}=\angle{QLC}=90$, thus $\angle{RPK}=\angle{RQL}=90+\dfrac{C}{2}$$\triangle{PKC} \sim \triangle{QLC}$, so $\dfrac{PK}{QL}=\dfrac{KC}{LC}$, or $\dfrac{PK}{QL}=\dfrac{BC}{AB}$. The ratio of the areas is $\dfrac{[RPK]}{[RQL]}=\dfrac{BC*RP}{AC*RQ}$. The two areas are only equal when the ratio is 1, therefore it suffices to show $\dfrac{RP}{RQ}=\dfrac{AC}{BC}$. Let $O$ be the center of the circle. Then $\angle{ROK}=A+C$, and $\angle{ROP}=180-(A+C)=B$. Using law of sines on $\triangle{RPO}$ we have: $\dfrac{RP}{\sin{B}}=\dfrac{OR}{\sin{(90+\dfrac{C}{2})}}$ so $RP*\sin{(90+\dfrac{C}{2})}=OR*\sin{B}$$OR*\sin{B}=\dfrac{1}{2}AC$ by law of sines, and $\sin{(90+\dfrac{C}{2})}=\cos{\dfrac{C}{2}}$, thus 1) $2RP\cos{\dfrac{C}{2}}=AC$. Similarly, law of sines on $\triangle{ROQ}$ results in $\dfrac{RQ}{\sin{(180-A)}}=\dfrac{OR}{\sin{(90-\dfrac{C}{2})}}$ or $\dfrac{RQ}{\sin{A}}=\dfrac{OR}{\cos{\dfrac{C}{2}}}$. Cross multiplying we have $RQ\cos{\dfrac{C}{2}}=OR*\sin{A}$ or 2) $2RQ\cos{\dfrac{C}{2}}=BC$. Dividing 1) by 2) we have $\dfrac{RP}{RQ}=\dfrac{AC}{BC}$

Problem 05

Let $n$ and $k$ be positive integers with $k \geq n$ and $k - n$ an even number. Let $2n$ lamps labelled $1$$2$, ..., $2n$ be given, each of which can be either on or off. Initially all the lamps are off. We consider sequences of steps: at each step one of the lamps is switched (from on to off or from off to on).

Let $N$ be the number of such sequences consisting of $k$ steps and resulting in the state where lamps $1$ through $n$ are all on, and lamps $n + 1$ through $2n$ are all off.

Let $M$ be number of such sequences consisting of $k$ steps, resulting in the state where lamps $1$ through $n$ are all on, and lamps $n + 1$ through $2n$ are all off, but where none of the lamps $n + 1$ through $2n$ is ever switched on.

Determine $\frac {N}{M}$.

Solution

For convenience, let $A$ denote the set $(1,2,\ldots n)$ and $B$ the set $(n+1,n+2,\ldots,2n)$.

We can describe each sequences of switching the lamps as a $k$-dimensional vector $(a_1, a_2, \ldots, a_k)$, where $a_i \in A \cup B$ signifies which lamp was switched on the $i$-th move for $i=1,2,\ldots k$.

Let $\cal{N}$ consist of those sequences that contain each of the numbers in $A$ an odd number of times and each of the numbers in $B$ an even number of times. Similarly, let $\cal{M}$ denote the set of those sequences that contain no numbers from $B$ and each of the numbers in $A$ an odd number of times. By definition, $M=|\cal{M}|$ and $N=|\cal{N}|$.

Define the mapping $f:\cal{N} \rightarrow \cal{M}$ as\[f(a_1, a_2, \ldots, a_k) = (b_1,b_2,\ldots b_k) :  b_i =  \begin{cases}    a_i, & \mbox{ if }  a_i \in A \\    a_i-n, & \mbox{ if } a_i  \in B \end{cases}\]

What we want to show now is that each element of $\cal{M}$ is an image of exactly $2^{k-n}$ elements from $\cal{N}$, which would imply $N = 2^{k-n}M$ and solve the problem.

Consider an arbitrary element $y$ of $\cal{M}$ and let $l_i$ be the number of appearances of the number $i$ in $y$ for $i=1,2,\ldots n$. Now consider the set of pre-images of $y$, that is $X_y = \{ x | f(x) = y \}$.

It is easy to see that each element $x\in X_y$ is derived from $y$ by flipping an even number of its $1$-s, $2$-s, and so on, where flipping means changing the number $j\in A$ to $j+n\in B$. Since each such set of flippings results in a unique $x$, all we want to count is the number of flippings. We can flip exactly $0, 2, 4,\ldots$ of the $1$-s, so that results in\[\binom{l_1}{0} + \binom{l_1}{2}+\binom{l_1}{4}+\cdots = 2^{l_1-1}\]flippings. Combine each of them with the $2^{l_2-1}$$2^{l_3-1}$, etc. ways of flipping the $2$-s, $3$-s etc. respectively to get the total number of flippings:\[2^{l_1-1}2^{l_2-1}\cdots2^{l_n-1} = 2^{l_1+l_2+\cdots+l_n-n} = 2^{k-n}.\]This shows that $|X_y| = 2^{k-n}$ and the proof is complete.

Problem 06

Let $n$ be a positive integer. Consider\[S=\{(x,y,z)~:~x,y,z\in \{0,1,\ldots,n \},~x+y+z>0\}\]as a set of $(n+1)^3-1$ points in three-dimensional space. Determine the smallest possible number of planes, the union of which contain $S$ but does not include $(0,0,0)$.

Solution

We will prove the result using the following Lemma, which has an easy proof by induction.

Lemma Let $S_1 = \{0, 1, \ldots, n_1\}$$S_2 = \{0, 1, \ldots, n_2\}$ and $S_3 = \{0, 1, \ldots, n_3\}$. If $P$ is a polynomial in $\mathbb{R}[x, y, z]$ that vanishes on all points of the grid $S = S_1 \times S_2 \times S_3$ except at the origin, then\[\deg P \geq n_1 + n_2 + n_3.\]

Proof. We will prove this by induction on $n = n_1 + n_2 + n_3$. If $n = 0$, then the result follows trivially. Say $n > 0$. WLOG, we can assume that $n_1 > 0$. By polynomial division over $\mathbb{R}[y, z][x]$ we can write\[P = (x - n_1)Q + R.\]Since $x-n_1$ is a monomial, the remainder $R$ must be a constant in $\mathbb{R}[y, z]$, i.e., $R$ is a polynomial in two variables $y$$z$. Pick an element of $S$ of the form $(n_1, y, z)$ and substitute it in the equation. Since $P$ vanishes on all such points, we get that $R(y, z) = 0$ for all $(y, z) \in S_2 \times S_3$. Let $S_1' = \{0, 1, \ldots, n_1 - 1\}$ and $S' = S_1' \times S_2 \times S_3$. For every point $(x, y, z)$ in $S'$ we have\[P(x, y, z) = (x - n_1)Q(x, y, z) + R(y, z) = (x-n_1)Q(x, y, z),\]where $x - n_1 \neq 0$. Therefore, the polynomial $Q$ vanishes on all points of $S'$ except the origin. By induction hypothesis, we must have $\deg Q \geq n_1 - 1 + n_2 + n_3$. But, $\deg P \geq \deg Q + 1$ and hence we have $\deg P \geq n_1 + n_2 + n_3$.


Now, to solve the problem let $H_1, \ldots, H_m$ be $m$ planes that cover all points of $S$ except the origin. Since these planes don't pass through origin, each $H_i$ can be written as $a_i x + b_i y + c_i z = 1$. Define $P$ to be the polynomial $\prod (a_ix + b_iy + c_iz - 1)$. Then $P$ vanishes at all points of $S$ except at the origin, and hence $\deg P = m \geq 3n$.

No comments:

Post a Comment