The Setting
Let , where is a finite index set. The question is whether the resolvent kernels of different effects can share a finite-dimensional purification space while retaining their mutual Gram operators.
An effect on is a positive semidefinite operator satisfying . Here is the identity on , denotes the Hermitian adjoint of , and is the Löwner order, meaning that is positive semidefinite.
Resolvent Kernels and Their Common Target
For a complex matrix on and a real parameter , the resolvent and purification kernel are the operators in (1). (1) For a complex matrix on , the effect entropy is the operator obtained by continuous functional calculus from the scalar function in (2), with the convention . (2)✦ All logarithms in (2) are natural.
The joint purification kernel of a finite family of complex matrices on , at a real parameter , is the block row operator defined in (3), where . (3)✦ For a finite family of complex matrices on , the common Gram operator on is the integral in (4). (4)✦ Let be the inclusion into the th summand. The finite common purification associated with a finite family of complex matrices on and an index is the column block of the positive square root of in (5). (5)✦
The Claim
Theorem 1. Every positive semidefinite operator on satisfies the normalization identity (6): (6)✦
Let be a finite family of effects on , and let satisfy . For an effect , the kernel distance bound is (7): (7)✦
For every finite family of positive semidefinite operators on , there are maps satisfying (8) for all : (8)✦
Suppose that every is an effect. Let be a finite set, , and . If , , and , the finite common purifications defined in (5) satisfy (9): (9)✦
In the paper, writing for the resolvent purification of an effect , the result reads: for every effect , ; for effects and positive weights summing to one, with , one has ; and every finite collection of effects admits a common finite-dimensional target for its resolvent maps preserving every product for effects and in that collection.
What Is Checked
The claims are stated over finite-dimensional complex matrices and their integrals for positive real ; the Jensen bounds use finite convex combinations of effects whose weights may vanish.