Difference between revisions of "2015 USAMO Problems/Problem 3"
m (→Solution) |
m (→Solution) |
||
Line 11: | Line 11: | ||
Let Sn={n} are one-element subsets. | Let Sn={n} are one-element subsets. | ||
− | Let mCk denote m choose k = m! | + | Let mCk denote m choose k = <math>\frac{m!}{k!(m-k)!}</math> |
(Case I) <math>f(\null)=1</math>. Then for distinct m and k, <math>f(Sm+Sk)=f(Sm)f(Sk)</math>, meaning only if Sm and Sk are both blue is their union blue. Namely <math>C(Sm+Sk)=C(Sm)C(Sk).</math> | (Case I) <math>f(\null)=1</math>. Then for distinct m and k, <math>f(Sm+Sk)=f(Sm)f(Sk)</math>, meaning only if Sm and Sk are both blue is their union blue. Namely <math>C(Sm+Sk)=C(Sm)C(Sk).</math> | ||
− | Similarly, for distinct m,n,k, f(Sm+Sk+Sn)=f(Sm+Sk)f(Sn), C(Sm+Sk+Sn)=C(Sm)C(Sk)C(Sn). This procedure of determination continues to S. Therefore, if T={a1,a2,...ak}, then C(T)=C(Sa1)C(Sa2)...C(Sak). All colorings thus determined by the free colors chosen for subsets of one single elements satisfy the condition. There are 2^n | + | Similarly, for distinct m,n,k, f(Sm+Sk+Sn)=f(Sm+Sk)f(Sn), C(Sm+Sk+Sn)=C(Sm)C(Sk)C(Sn). This procedure of determination continues to S. Therefore, if T={a1,a2,...ak}, then C(T)=C(Sa1)C(Sa2)...C(Sak). All colorings thus determined by the free colors chosen for subsets of one single elements satisfy the condition. There are 2^n colorings in this case. |
− | (Case II.) f(<math>\ | + | (Case II.) f(<math>\varnothing</math>)=0. |
− | (Case II.1) Core=<math>\ | + | (Case II.1) Core=<math>\varnothing</math>. Then either (II.1.1) there exist two nonintersecting subsets A and B, C(A)=C(B)=1, but f(A)f(B)=0 which is a contradiction, or (II.1.2) all subsets has C(T)=0, which is easily confirmed to satisfy the condition f(T1)f(T2)=f(T1^T2)f(T1+T2). There is one legitimate coloring in this case. |
(Case II.2) Core= a subset of 1 element. WOLG, C(S1)=1. Then f(S1)=1, and subsets containing element 1 may be colored Blue. | (Case II.2) Core= a subset of 1 element. WOLG, C(S1)=1. Then f(S1)=1, and subsets containing element 1 may be colored Blue. | ||
− | <math>f(S1+Sm)f(S1+Sn)=f(S1+Sm+Sn)</math>, namely C(S1+Sm+Sn)=C(Sm+S1)C(Sn+S1). Now S1 functions as the | + | <math>f(S1+Sm)f(S1+Sn)=f(S1+Sm+Sn)</math>, namely C(S1+Sm+Sn)=C(Sm+S1)C(Sn+S1). Now S1 functions as the <math>\varnothing</math> in case I, with n-1 elements to combine into a base of n-1 2-element sets, and all the other subsets are determined. There are 2^(n-1) legit colorings for each choice of core. But there are nC1 (i.e. n choose 1) = n such cores. Hence altogether there are n2^(n-1) colorings in this case. |
(Case II.3) Core = a subset of 2 elements. WLOG, C(S1+S2)=1. Only subsets containing the core may be colored Blue. With the same reasoning as in the preceding case, there are (nC2)2^(n-2) colorings. | (Case II.3) Core = a subset of 2 elements. WLOG, C(S1+S2)=1. Only subsets containing the core may be colored Blue. With the same reasoning as in the preceding case, there are (nC2)2^(n-2) colorings. |
Revision as of 20:09, 16 January 2017
Problem
Let , where . Each of the subsets of is to be colored red or blue. (The subset itself is assigned a color and not its individual elements.) For any set , we then write for the number of subsets of T that are blue.
Determine the number of colorings that satisfy the following condition: for any subsets and of ,
Solution
Define function: if the set T is colored blue, and if is colored red. Define the .
The empty set is denoted as , ^ denotes intersection; + denotes union. T1<T2 means T1 is a subset of T2 but not =T2. Let Sn={n} are one-element subsets.
Let mCk denote m choose k =
(Case I) . Then for distinct m and k, , meaning only if Sm and Sk are both blue is their union blue. Namely
Similarly, for distinct m,n,k, f(Sm+Sk+Sn)=f(Sm+Sk)f(Sn), C(Sm+Sk+Sn)=C(Sm)C(Sk)C(Sn). This procedure of determination continues to S. Therefore, if T={a1,a2,...ak}, then C(T)=C(Sa1)C(Sa2)...C(Sak). All colorings thus determined by the free colors chosen for subsets of one single elements satisfy the condition. There are 2^n colorings in this case.
(Case II.) f()=0.
(Case II.1) Core=. Then either (II.1.1) there exist two nonintersecting subsets A and B, C(A)=C(B)=1, but f(A)f(B)=0 which is a contradiction, or (II.1.2) all subsets has C(T)=0, which is easily confirmed to satisfy the condition f(T1)f(T2)=f(T1^T2)f(T1+T2). There is one legitimate coloring in this case.
(Case II.2) Core= a subset of 1 element. WOLG, C(S1)=1. Then f(S1)=1, and subsets containing element 1 may be colored Blue.
, namely C(S1+Sm+Sn)=C(Sm+S1)C(Sn+S1). Now S1 functions as the in case I, with n-1 elements to combine into a base of n-1 2-element sets, and all the other subsets are determined. There are 2^(n-1) legit colorings for each choice of core. But there are nC1 (i.e. n choose 1) = n such cores. Hence altogether there are n2^(n-1) colorings in this case.
(Case II.3) Core = a subset of 2 elements. WLOG, C(S1+S2)=1. Only subsets containing the core may be colored Blue. With the same reasoning as in the preceding case, there are (nC2)2^(n-2) colorings.
... (Case II.n+1) Core =S. Then C(S)=1, with all other subsets C(T)=0, there is 1=(nCn)2^0
Combining all the cases, is the total number of colorings satisfying the given condition.