fra 🥀 :
personally, i really like the proof using minimal polynomials, because it shows how linear algebra can actually be /algebra/, as in abstract algebra, and not just computations with matrices. it uses a beautiful theorem that generalizes the chinese remainder theorem, saying that if a polynomial p can be factored into pairwise coprime factors q_1,...,q_r, and A is a square matrix such that p(A)=0, then V is the direct sum of all the Ker(q_i(A)).
in this case, if A^k=id, it means that its minimal polynomial divides x^k-1. if k is prime to the characteristic of the field, this implies that the minimal polynomial is separable, so (assuming the field is algebraically closed) the minimal polynomial looks like (x-a_1)*...*(x-a_r). by the previous theorem, this means that V is the direct sum of the eigenspaces of A, meaning that A is diagonalizable.
2025-07-22 09:07:38