The Setting
Consider repeated uses of a generalized amplitude-damping channel with damping and stationary excited-state population . Separating the first qubit of a pure input gives two amplitude branches on the remaining qubits.
The computational basis of qubits is the set of bit strings in (1), where .✦ An -qubit operator is a complex matrix with rows and columns indexed by .✦ (1)✦
The single-qubit Kraus operators are the matrices in (2). (2)✦
For a label , the tensor product Kraus operator is the -qubit operator with entries given by (3). (3)✦ The memoryless channel is the map on -qubit operators defined in (4), with denoting the adjoint. (4)✦
For complex amplitudes on a finite basis , the squared norm and pure state operator are defined by (5). (5) Fix a reference element . The reference basis vector and normalized amplitude vector for complex amplitudes on the finite basis are the vectors on given by (6). (6)
The branch amplitudes of an -qubit amplitude vector , for , are its restrictions in (7); denotes the bit string with first bit and tail . (7)✦ The excitation population at site of amplitudes on is the sum in (8). (8)✦
The entropy of a finite complex square matrix is the real trace of the spectral functional calculus for , as in (9).✦ Here , and entropies are measured in nats. (9)✦ The qubit entropy function and output determinant are defined by (10), where is binary entropy with natural logarithms. (10)
The Claim
Theorem 1. Let , let be a nonnegative integer, and let be amplitudes on with . Put and for . The output entropies satisfy (11). (11)✦
In the paper, the result reads: for , put and ; if is a completely positive trace-preserving map between operators on nonzero finite-dimensional complex Hilbert spaces, where denotes the space of linear operators on , and is a unit vector in , with and unit branch vectors chosen arbitrarily at zero weight, then .
What Is Checked
The claim is stated over pure inputs of squared norm one and powers of a fixed generalized amplitude-damping channel at parameters , with branch normalization as in (6).