uniolympiads English

IMC 2022, Ημέρα 2, Θέμα 7

Let A1,A2,…,AkA_1, A_2, \dots, A_k be n×nn \times n idempotent complex matrices such that

AiAj=−AjAifor all i≠j.A_i A_j = -A_j A_i \quad \text{for all } i \ne j.

Prove that at least one of the given matrices has rank ≤nk\le \frac{n}{k}. (A matrix AA is called idempotent if A2=AA^2 = A.)

Πρωτότυπο από το επίσημο PDFLet A 1 , A 2 , . . . , A k be n × n idempotent complex matrices such that A i A j = −A j A i for all i ̸ = j. Prove that at least one of the given matrices has rank ≤ nk . (A matrix A is called idempotent if A 2 = A.)

Η μεταγραφή συγκρίθηκε σύμβολο προς σύμβολο με το πρωτότυπο, όχι ακόμα από άνθρωπο.

Χρόνος σε αυτό το θέμα 0:00

Δούλεψε τουλάχιστον 45 λεπτά πριν ανοίξεις τη λύση.

Επίσημη λύση

(proposed by Danila Belousov, Novosibirsk)

Hint: Consider the trace and the rank of AA.

Solution 1.

Lemma. For any idempotent matrix BB

tr(B)=rank(B)\mathrm{tr}(B) = \mathrm{rank}(B)

Proof. Observe that an idempotent matrix satisfies the equation λ(1−λ)=0\lambda(1-\lambda) = 0. Hence the minimal polynomial is a product of linear factors and the matrix is diagonalizable. Therefore, the rank of the matrix equals the number of non-zero eigenvalues. Since the matrix has eigenvalues 0 or 1, this provides that the trace is equal to the number of unity eigenvalues, or non-zero eigenvalues.

It can be shown that ∑i=1kAi\sum\limits_{i=1}^{k} A_i is also an idempotent. Indeed,

(∑i=1kAi)2=∑i=1kAi2+∑i≠j(AiAj+AjAi)=∑i=1kAi\left(\sum_{i=1}^{k} A_i\right)^2 = \sum_{i=1}^{k} A_i^2 + \sum_{i \ne j} (A_i A_j + A_j A_i) = \sum_{i=1}^{k} A_i

Applying the lemma one can obtain

∑i=1krank(Ai)=∑i=1ktr(Ai)=tr(∑i=1kAi)=rank(∑i=1kAi)⩽n\sum_{i=1}^{k} rank(A_i) = \sum_{i=1}^{k} tr(A_i) = tr\left(\sum_{i=1}^{k} A_i\right) = rank\left(\sum_{i=1}^{k} A_i\right) \leqslant n

The required inequality follows.

Solution 2. We first prove that for idempotents A,BA, B with AB=−BAAB = -BA we already must have AB=BA=0AB = BA = 0. Indeed, it is clear that ABx=BAx=0ABx = BAx = 0 for x∈ker⁡(A)x \in \ker(A) so it suffices to prove the same for x∈im(A)x \in \mathrm{im}(A), i.e. when Ax=xAx = x. But then writing Bx=yBx = y we have Ay=−yAy = -y i.e. y=−Ay=−A2y=Ay=−yy = -Ay = -A^2y = Ay = -y and hence y=0y = 0 so that again ABx=BAx=0ABx = BAx = 0.

Henceforth, we can assume the stronger condition AiAj=0A_iA_j = 0 for all i≠ji \ne j. We next claim that all the image spaces ViV_i of AiA_i are linearly independent. This will imply the claim, since then the sum of their dimensions can be at most nn, and so one of them has to be ≤nk\le \frac{n}{k}. Now, for the sake of contradiction, suppose that ∑ivi=0\sum_i v_i = 0 with vi∈Viv_i \in V_i and w.l.o.g. v1≠0v_1 \ne 0. But then

0=A1(v1+⋯+vk)=v1+A1v2+⋯+A1vk=v1+A1A2v2+⋯+A1Akvk=v10 = A_1(v_1 + \cdots + v_k) = v_1 + A_1v_2 + \cdots + A_1v_k = v_1 + A_1A_2v_2 + \cdots + A_1A_kv_k = v_1
since A1Ai=0A_1A_i = 0 for all ii.

Remark. Here is a different argument for AB=BA=0AB = BA = 0, without eigenvectors: multiplying by AA and using its idempotence and the super-commutativity , we have

−BA=AB=A2B=AAB=−ABA=BAA=BA2=BA-BA = AB = A^2B = AAB = -ABA = BAA = BA^2 = BA
thus BA=0BA = 0.

Μεταγραφή από το επίσημο κείμενο λύσεων, σελίδες 3–4. Η μεταγραφή δεν έχει ελεγχθεί δεύτερη φορά· αν έχεις αμφιβολία, κοίτα το πρωτότυπο.

Θέμα αρχείου: η εκφώνηση είναι το επίσημο κείμενο (μεταγραμμένο, με το πρωτότυπο παραπάνω). Η ταξινόμηση ανά αντικείμενο και η δυσκολία εκτιμήθηκαν αυτόματα· υποδείξεις και λύση στα ελληνικά προστίθενται σταδιακά.

Πώς πήγε;