Majorization in C*-Algebras
The notion of majorization of one self-adjoint n×n matrix by another is a very useful concept in linear algebra. For example, a classical theorem of Schur and Horn states that a diagonal matrix D is majorized by a self-adjoint matrix B if and only if a unitary conjugate of B has the same diagonal as D. Some equivalent characterizations of A being majorized by B include A being in the convex hull of the unitary orbit of B (that is, there exists a mixed unitary quantum channel that maps B to A), the eigenvalues of A being controlled by the eigenvalues of B via certain inequalities, tracial inequalities involving convex functions of A and B, and doubly stochastic matrices relating the eigenvalues of A and B.
In this talk, we will examine the notion of majorization in unital C∗-algebras. In particular, the closed convex hulls of the unitary orbits of a self-adjoint operator will be characterized, the possible values of the expectation of an operator onto a MASA in a II1 factor will be described, and generalizations to (commuting) tuples of self-adjoint operators will be discussed.