Unofficial archive — problems, solutions & results © IMC, reproduced with permission.

IMC / 2004 / Problems / Day 2, P12

IMC 2004 · Day 2 · P12

killer

For n0n \ge 0 define matrices AnA_n and BnB_n as follows: A0=B0=(1)A_0 = B_0 = (1) and for every n>0n > 0 An=(An1An1An1Bn1)andBn=(An1An1An10).A_n = \begin{pmatrix} A_{n-1} & A_{n-1} \\ A_{n-1} & B_{n-1} \end{pmatrix} \quad \text{and} \quad B_n = \begin{pmatrix} A_{n-1} & A_{n-1} \\ A_{n-1} & 0 \end{pmatrix}. Denote the sum of all elements of a matrix MM by S(M)S(M). Prove that S(Ank1)=S(Akn1)S(A_n^{k-1}) = S(A_k^{n-1}) for every n,k1n, k \ge 1.

Solution (official)

The quantity S(Ank1)S(A_n^{k-1}) has a special combinatorical meaning. Consider an n×kn \times k table filled with 0's and 1's such that no 2×22 \times 2 contains only 1's. Denote the number of such fillings by FnkF_{nk}. The filling of each row of the table corresponds to some integer ranging from 0 to 2n12^n - 1 written in base 2. FnkF_{nk} equals to the number of kk-tuples of integers such that every two consecutive integers correspond to the filling of n×2n \times 2 table without 2×22 \times 2 squares filled with 1's.

Consider binary expansions of integers ii and jj: inin1i1i_n i_{n-1} \dots i_1 and jnjn1j1j_n j_{n-1} \dots j_1. There are two cases:

1. If injn=0i_n j_n = 0 then ii and jj can be consecutive iff in1i1i_{n-1} \dots i_1 and jn1j1j_{n-1} \dots j_1 can be consequtive.

2. If in=jn=1i_n = j_n = 1 then ii and jj can be consecutive iff in1jn1=0i_{n-1} j_{n-1} = 0 and in2i1i_{n-2} \dots i_1 and jn2j1j_{n-2} \dots j_1 can be consecutive.

Hence ii and jj can be consecutive iff (i+1,j+1)(i+1, j+1)-th entry of AnA_n is 1. Denoting this entry by ai,ja_{i,j}, the sum S(Ank1)=i1=02n1ik=02n1ai1i2ai2i3aik1ikS(A_n^{k-1}) = \sum_{i_1=0}^{2^n-1} \dots \sum_{i_k=0}^{2^n-1} a_{i_1 i_2} a_{i_2 i_3} \dots a_{i_{k-1} i_k} counts the possible fillings. Therefore Fnk=S(Ank1)F_{nk} = S(A_n^{k-1}).

The the obvious statement Fnk=FknF_{nk} = F_{kn} completes the proof.

How the field did

contestants scored
176
average (of 20)
0.45
solved (≥ 80%)
0.6%
near-0 (≤ 10%)
97.2%
discrimination
0.13

Score distribution (field cohort)

Computed on contestants with a meaningful total (field cohort); discrimination is the corrected item–total correlation.

Similar problems

IMC 2021 · Day 2 · P8killeravg 0.0/10 · solved 0% · near-0 99% · disc 0.08
IMC 2010 · Day 2 · P9killeravg 0.5/10 · solved 2% · near-0 93% · disc 0.42
IMC 2024 · Day 1 · P3mediumavg 5.3/10 · solved 40% · near-0 30% · disc 0.55