A note on \(k\)-potence preserving maps (Q947621): Difference between revisions

Let \(M_n\) be the algebra of all \(n\times n\) matrices and \(\Gamma_n\) the set of all \(k\)-potent matrices in \(M_n\) (i.e., \(A\in M_n\) is said to be \(k\)-potent if \(A^k=A\)). The paper proves the following result: Suppose \(\varphi:M_n\to M_n\) is a map satisfying \(A-\lambda B\in\Gamma_n\) implies \(\varphi(A)- \lambda\varphi(B)\in\Gamma_n\), where \(A,B\in M_n\), \(\lambda\in\mathbb C\). Then either \(\varphi\) is of the form \(\varphi(A)= cTAT^{-1}\), \(A\in M_n\), or \(\varphi\) is of the form \(\varphi(A)= cTA^tT^{-1}\), \(A\in M_n\), where \(T\in M_n\) is an invertible matrix, \(c\in\mathbb C\) satisfies \(c^k=c\).