This computer-assisted preprint proves a sharp reverse relative entropy inequality L >= beta_m chi for uniform ensembles of individually normalized qubit states, for every m >= 2. Here chi is the Holevo quantity and L is the non-optimized quantum lautum information. The proof combines analytic Bloch-vector and Gram-matrix reductions with four interval certificates covering 855,780 boxes. For three matrices in arbitrary finite dimension, the paper also establishes local nonnegativity near faithful pairwise-commuting equality triples, allowing unequal traces and noncommuting perturbations. Further results include normalization bridges, an angular reduction for three qubits, and certified obstructions to proposed pinching and filtering descent arguments. The unrestricted unequal-trace ensemble inequality and the general equality of complete and ordinary modified logarithmic Sobolev constants are not proved. The accompanying archive contains sources, certificate inputs, verifiers, and reproduced results. Developed with AI assistance; not peer reviewed or formally verified. Companion paper: Christoph Hartmann, Reverse relative entropy and finite-time contraction for conditional expectations (2026).
This computer-assisted preprint proves a sharp reverse relative entropy inequality L >= beta_m chi for uniform ensembles of individually normalized qubit states, for every m >= 2. Here chi is the Holevo quantity and L is the non-optimized quantum lautum information. The proof combines analytic Bloch-vector and Gram-matrix reductions with four interval certificates covering 855,780 boxes. For three matrices in arbitrary finite dimension, the paper also establishes local nonnegativity near faithful pairwise-commuting equality triples, allowing unequal traces and noncommuting perturbations. Further results include normalization bridges, an angular reduction for three qubits, and certified obstructions to proposed pinching and filtering descent arguments. The unrestricted unequal-trace ensemble inequality and the general equality of complete and ordinary modified logarithmic Sobolev constants are not proved. The accompanying archive contains sources, certificate inputs, verifiers, and reproduced results. Developed with AI assistance; not peer reviewed or formally verified. Companion paper: Christoph Hartmann, Reverse relative entropy and finite-time contraction for conditional expectations (2026).
This preprint proves reverse relative entropy and finite-time contraction inequalities for trace-preserving conditional expectations on finite-dimensional matrix algebras. A spectral-block reduction for an algebra-valued multiplier and a scalar boundary reduction yield explicit constants gamma_C and Lambda_p(C). Complete bounds follow from complete order bounds. For qubit depolarization, the results give 0.940046944825 <= alpha_c < 0.954706390024 in the convention exp(-2 alpha t), and 0.265671889879 < eta_c < 0.270961852526 at retained weight p=1/2. The contraction coefficient compares each state with its own conditional-expectation equilibrium. The paper also proves a stronger bound for commuting states, with sharpness in specified classes, and derives consequences for products of local channels. The exact complete qubit constants and the unrestricted sharp ensemble inequality are not determined here. The accompanying archive contains LaTeX sources, exact inputs, symbolic and interval verifiers, numerical regression checks, and reproduced outputs. Developed with AI assistance; not peer reviewed or formally verified. Companion paper: Christoph Hartmann, Sharp reverse entropy inequalities for uniform qubit ensembles (2026).
Hirche and Reeb conjectured a sharp erasure-channel upper bound for the entropy of the XOR of independent uniform bits with arbitrary quantum side information. We establish this bound for a noncommuting class of binary classical–quantum ensembles: one average-state/signal pair admits an orthogonal decomposition into signed rank-one signal blocks, while the other acts on a qubit with an arbitrary traceless Hermitian signal. The principal analytic ingredient is a Schur-convexity theorem for symmetrized even-order qubit resolvent trace kernels. Combined with a positive-association principle for uniform Dirichlet vectors under majorization, this yields tensor-product resolvent-moment inequalities at every even order. A positive entropy-moment representation and Chebyshev association then establish the sharp conditional-entropy inequality. Additional corollaries cover two arbitrary-dimensional rank-one-block ensembles and gauge-balanced higher-rank signal blocks. The manuscript provides complete analytical proofs, including the simplex association and trace-kernel results, together with reproducible deterministic verification of the finite trace algebra and numerical tests on physically admissible noncommuting quantum ensembles. The results establish the sharp Hirche–Reeb information-combining inequality for the stated structural classes. The unrestricted conjecture for arbitrary quantum side information remains open.
Given a finite-range, non-commuting Hamiltonian on a 1D chain of finite spins, we prove a modified logarithmic Sobolev inequality with constant Ω(1/logn) for the heat-bath dynamics, a quasi-local Gibbs sampler, and for the regularised heat-bath dynamics. This implies rapid mixing of both samplers towards the Gibbs state in relative entropy and trace norm, with mixing times of order O(logn⋅log(n/ε)). The interactions may be spatially inhomogeneous. The proof combines weak quasi-factorisation of relative entropy, decay-of-correlation estimates, a uniform conditional local gap, and a buffered comparison between conditional entropy and quadratic energy. The resulting bounded entropy defect is removed using the global gap. To the best of our knowledge, this is the first proof of MLSI in the non-commuting regime.
This paper presents self-contained proofs of the strong subadditivity inequality for quantum entropy and some related inequalities for the quantum relative entropy, most notably its convexity and its monotonicity under stochastic maps. Moreover, the approach presented here, which is based on Klein's inequality and one of Lieb's less well-known concave trace functions, allows one to obtain conditions for equality. Using the fact that the Holevo bound on the accessible information in a quantum ensemble can be obtained as a consequence of the monotonicity of relative entropy, we show that equality can be attained for that bound only when the states in the ensemble commute. The paper concludes with an Appendix giving a short description of Epstein's elegant proof of the relevant concavity theorem of Lieb.
How much distinguishability can a noisy quantum channel preserve when its input is correlated with an untouched quantum memory? This paper gives an exact finite semidefinite program for that question for every finite-dimensional completely positive, trace-preserving channel. The complete coefficient compares joint states with the same memory marginal. The conditional coefficient instead subtracts the distinguishability already present in the memory. Both coefficients, and conditional mutual-information contraction with a separate conditioning system, equal one minus the reciprocal of an optimization in the channel’s unnormalized Choi matrix. The equivalent inverse formula retains an essential range condition, including for singular channels. The proof solves a centered matrix-Gram problem, shows that its optimization center can be positive, and constructs a signed observable lift through an erasure channel. Two weighted quadratic inequalities give the universal entropy bound. A rare coherent memory sector proves sharpness while preserving both local marginals exactly. A memory of bounded dimension suffices, including with prescribed faithful input marginals. The same program gives exact coefficients for regular modular Petz divergences whose generators have finite endpoint value and recession slope, every fixed Petz Rényi order strictly between zero and one, and quantum Chernoff divergence. Further results include a harmonic dual, an exact tensor-product law, a strict-contraction criterion, a sharp comparison with positive quantum Doeblin minorization, and explicit classical, depolarizing and orthogonal-unitary-noise formulas. Ordinary quantum mutual-information contraction is a different optimization. For qubit depolarization the paper determines its exact value on the full completely positive parameter interval. At shrinkage one half, ordinary relative-entropy, ordinary mutual-information and complete entropy contraction are respectively 1/4, 1/3 and 3/8. In higher dimension, bounds and a strict separation are proved without asserting an exact ordinary-MI formula. A discontinuity between nearly complete amplitude damping and pure reset rules out uniformly bounded-use estimation of the exact coefficient from an unknown channel, even with adaptive experiments and quantum ancillas. Thus evaluation from exact supplied Choi data does not imply stable experimental acquisition. Rare sharpness witnesses can carry vanishing absolute information; the results do not assert a communication rate, physical simulator, general recovery map or polynomial cost in physical qubit count.
We determine the complete equal-reference Umegaki relative-entropy contraction coefficient of every finite preparation channel after input dephasing. For prepared states ρ₁, …, ρₙ, let H be the intersection of their supports and let Aᵢ be the compression of ρᵢ⁺ to H. If H is nonzero, the coefficient is 1 − min over p in Δₙ of Tr_H[(Σᵢ pᵢAᵢ)⁻¹]; it equals one when H is zero. A noncommutative metric inequality gives the upper bound, and a weighted operator-simplex construction gives matching faithful classical–quantum input pairs with equal classical and reference marginals. Reference dimension n suffices for the supremum. The complementary coefficient is the reciprocal of an exponential left order-two Petz–Rényi radius and is multiplicative under Cartesian products. Faithful binary-input qubit-output channels can consequently have complete coefficient tending to one while their ordinary coefficient tends to zero, with reference dimension two sufficient throughout. An exact rational example has complete coefficient 9/41. The matching input family converges pointwise, has input relative entropy tending to zero, and can converge logarithmically slowly; it provides no fixed-error or capacity conclusion. A visual guide records the mathematical routes, proof dependencies, and provenance of the result.