Barycentric Rényi divergences were introduced in [Mosonyi, Bunth, Vrana, Linear Algebra and its Applications, 2024] as an alternative to standard Kubo-Ando constructions to define multivariate quantum Rényi divergences. They are defined via a variational expression and depend on a finite collection of quantum relative entropies $D^{q_x}$. When all the relative entropies are monotone under CPTP maps then so are the corresponding barycentric Rényi divergences, and when all the relative entropies are additive then the corresponding barycentric Rényi divergences are subadditive under tensor product. Additivity has only been established before for the case where all $D^{q_x}$ are chosen to be the Umegaki relative entropy, which is also the only case where the barycentric Rényi divergence (called the minimal one) was known to admit an explicit expression.
Here we settle the problem of additivity by showing that for any choice of monotone quantum relative entropies, the regularized barycentric Rényi divergence coincides with the minimal barycentric Rényi divergence on invertible inputs. This in turn implies that if at least two of the $D^{q_x}$ are strictly larger than the Umegaki relative entropy (in a precise sense), then the resulting barycentric Rényi divergence is not even weakly additive.
AI Summary
Barycentric Rényi divergences extend classical multivariate Rényi divergences through $D_P^{\mathrm{bary},q}(W)=\inf_\omega\sum_xP(x)D^{q_x}(\omega\|W_x)$. The paper addresses whether these quantities inherit tensor-product additivity from their generating quantum relative entropies, and computes their asymptotic regularizations.
For any collection of CPTP-monotone generating relative entropies and invertible inputs, it proves $\lim_{n\to\infty}n^{-1}D_P^{\mathrm{bary},q}(W^{\otimes n})=-\log\operatorname{Tr}\exp(\sum_xP(x)\log W_x)$, independently of the generators. For singular inputs, the lower and upper regularizations lie between the Umegaki barycentric divergence and an explicit log-Euclidean expression using the absolutely continuous parts of the inputs on their joint support. These bounds coincide when that support commutes with every relevant input. The measured construction attains the lower endpoint asymptotically; geometric-Umegaki constructions, including the Belavkin–Staszewski case, attain the upper endpoint.
The proof combines measured/maximal relative-entropy bounds with explicitly constructed trial states obtained by geometrically tilting the Umegaki optimizer toward universal symmetric states. It gives fixed-input convergence bounds of order $\sqrt{\log(n+1)/n}$. Under lower semicontinuity, weak subadditivity, and strict domination of Umegaki entropy for noncommuting pairs by at least two generators, suitable invertible inputs exhibit a strict single-copy/regularized gap, ruling out weak additivity.
General single-copy formulas remain unavailable outside two identified constructions. For arbitrary singular inputs and arbitrary monotone generators, the paper provides bounds rather than a universal regularization formula.
No sign of AI-generated textAI use disclosed
ChatGPT-6 Astra was instrumental in obtaining the results of the paper.
Error found: The abstract omits the weak subadditivity and lower semicontinuity assumptions under which the paper proves its non-weak-additivity consequence.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
Barycentric Rényi divergences were introduced in [Mosonyi, Bunth, Vrana, Linear Algebra and its Applications, 2024] as an alternative to standard Kubo-Ando constructions to define multivariate quantum Rényi divergences. They are defined via a variational expression and depend on a finite collection of quantum relative entropies Dqx. When all the relative entropies are monotone under CPTP maps then so are the corresponding barycentric Rényi divergences, and when all the relative entropies are additive then the corresponding barycentric Rényi divergences are subadditive under tensor product. Additivity has only been established before for the case where all Dqx are chosen to be the Umegaki relative entropy, which is also the only case where the barycentric Rényi divergence (called the minimal one) was known to admit an explicit expression.
Here we settle the problem of additivity by showing that for any choice of monotone quantum relative entropies, the regularized barycentric Rényi divergence coincides with the minimal barycentric Rényi divergence on invertible inputs. This in turn implies that if at least two of the Dqx are strictly larger than the Umegaki relative entropy (in a precise sense), then the resulting barycentric Rényi divergence is not even weakly additive.
Barycentric Rényi divergences extend classical multivariate Rényi divergences through DPbary,q(W)=infω∑xP(x)Dqx(ω∥Wx). The paper addresses whether these quantities inherit tensor-product additivity from their generating quantum relative entropies, and computes their asymptotic regularizations.
For any collection of CPTP-monotone generating relative entropies and invertible inputs, it proves limn→∞n−1DPbary,q(W⊗n)=−logTrexp(∑xP(x)logWx), independently of the generators. For singular inputs, the lower and upper regularizations lie between the Umegaki barycentric divergence and an explicit log-Euclidean expression using the absolutely continuous parts of the inputs on their joint support. These bounds coincide when that support commutes with every relevant input. The measured construction attains the lower endpoint asymptotically; geometric-Umegaki constructions, including the Belavkin–Staszewski case, attain the upper endpoint.
The proof combines measured/maximal relative-entropy bounds with explicitly constructed trial states obtained by geometrically tilting the Umegaki optimizer toward universal symmetric states. It gives fixed-input convergence bounds of order log(n+1)/n. Under lower semicontinuity, weak subadditivity, and strict domination of Umegaki entropy for noncommuting pairs by at least two generators, suitable invertible inputs exhibit a strict single-copy/regularized gap, ruling out weak additivity.
General single-copy formulas remain unavailable outside two identified constructions. For arbitrary singular inputs and arbitrary monotone generators, the paper provides bounds rather than a universal regularization formula.
Barycentric Rényi divergences were introduced in [Mosonyi, Bunth, Vrana, Linear Algebra and its Applications, 2024] as an alternative to standard Kubo-Ando constructions to define multivariate quantum Rényi divergences. They are defined via a variational expression and depend on a finite collection of quantum relative entropies Dqx. When all the relative entropies are monotone under CPTP maps then so are the corresponding barycentric Rényi divergences, and when all the relative entropies are additive then the corresponding barycentric Rényi divergences are subadditive under tensor product. Additivity has only been established before for the case where all Dqx are chosen to be the Umegaki relative entropy, which is also the only case where the barycentric Rényi divergence (called the minimal one) admits an explicit expression. Here we settle the problem of additivity by showing that for any choice of additive and monotone quantum relative entropies, the regularized barycen Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Quantum generalizations of Rényi's entropies are a useful tool to describe a variety of operational tasks in quantum information processing. Two families of such generalizations turn out to be particularly useful: the Petz quantum Rényi divergence Dˉα and the minimal quantum Rényi divergence Dα. Moreover, the maximum quantum Rényi divergence Dα is of particular mathematical interest. In this Master thesis, we investigate relations between these divergences and their applications in quantum information theory. Our main result is a reverse Araki-Lieb-Thirring inequality that implies a new relation between the minimal and the Petz divergence, namely that αDˉα(ρ∥σ)⩽Dα(ρ∥σ) for α∈[0,1] and where ρ and σ are density operators. This bound suggests defining a "pretty good fidelity", whose relation to the usual fidelity implies the known relations between the optimal and pretty good measurement as well as the optimal and pretty good singlet fraction. In addition, we provide a new proof of the inequality D1(ρ∥σ)⩽D1(ρ∥σ), based on the Araki-Lieb-Thirring inequality. This leads to an elegant proof of the logarithmic form of the reverse Golden-Thompson inequality.
Most quantum divergences derive their structure from classical f-divergences or Renyi-type constructions, a dependence that obscures several quantum geometric effects. We introduce a quantum relative-alpha-entropy that extends Umegaki's relative entropy while falling outside the f-divergence class. The proposed divergence exhibits a nonlinear convexity property, which yields a generalized convexity result for the Petz-Renyi divergence for alpha greater than one, complementing the known convexity for alpha less than one. It is additive under tensor products, invariant under unitary transformations, and depends only on the relative geometry of quantum states rather than their absolute magnitudes. Using Nussbaum-Szkola-type distributions, we also establish an exact correspondence of this divergence with classical relative-alpha-entropy. This reveals relative-alpha-entropy as a fundamentally geometric notion of quantum distinguishability not captured by existing divergence frameworks.
Abstract The asymptotic restriction problem for tensors can be reduced to finding all parameters that are normalized, monotone under restrictions, additive under direct sums and multiplicative under tensor products, the simplest of which are the flattening ranks. Over the complex numbers, a refinement of this problem, originating in the theory of quantum entanglement, is to find the optimal rate of entanglement transformations as a function of the error exponent. This trade-off can also be characterized in terms of the set of normalized, additive, multiplicative functionals that are monotone in a suitable sense, which includes the restriction-monotones as well. For example, the flattening ranks generalize to the (exponentiated) Rényi entanglement entropies of order α∈[0,1] α ∈ [ 0 , 1 ] . More complicated parameters of this type are known, which interpolate between the flattening ranks or Rényi entropies for special bipartitions, with one of the parts being a single tensor factor. We introduce a new construction of subadditive and submultiplicative monotones in terms of a regularized Rényi divergence between many copies of the pure state represented by the tensor and a suitable sequence of positive operators. We give explicit families of operators that correspond to the flattening-based functionals, and show that they can be combined in a nontrivial way using weighted operator geometric means. This leads to a new characterization of the previously known additive and multiplicative monotones, and gives new submultiplicative and subadditive monotones that interpolate between the Rényi entropies for all bipartitions. We show that for each such monotone there exist pointwise smaller multiplicative and additive ones as well. In addition, we find lower bounds on the new functionals that are superadditive and supermultiplicative.
We introduce a new cumulant-based quantum relative Rényi functional as a candidate quantum Rényi divergence, derived from the cumulant-generating function (CGF) of the quantum relative surprisal operator and extending the classical connection between Rényi divergence and statistical cumulants to the quantum setting. Unlike the Petz and sandwiched quantum Rényi divergences, the proposed construction is motivated by statistical structure rather than operator-algebraic or operational principles. The functional naturally admits a path-integral-like representation through the Lie-Trotter product expansion, providing a trajectory-based interpretation of quantum divergence in Hilbert space. On its natural non-regularized domain for α>1 under the support condition supp(ρ)⊆supp(σ), we establish several fundamental properties, including positivity, reduction to the classical case, additivity, unitary invariance, continuity, and monotonicity with respect to the Renyi parameter α. Whether the functional satisfies the quantum data-processing inequality (QDPI) under arbitrary CPTP maps remains open. To extend the analysis beyond the studied regime, we introduce a regularized version of the functional and study its behavior at α=0. We show that the resulting relative quantumness quantity vanishes if and only if the underlying states commute, yielding a necessary and sufficient characterization of non-commutativity. For commutativity-preserving (CoP) channels, we further conjecture a QDPI-type monotonicity relation for this quantity. Extensive numerical simulations provide strong evidence in support of this conjecture, with no violations observed for the CoP channels considered in this work.
The concept of classical [Formula: see text]-divergences gives a unified framework to construct and study measures of dissimilarity of probability distributions; special cases include the relative entropy and the Rényi divergences. Various quantum versions of this concept, and more narrowly, the concept of Rényi divergences, have been introduced in the literature with applications in quantum information theory; most notably Petz’ quasi-entropies (standard [Formula: see text]-divergences), Matsumoto’s maximal [Formula: see text]-divergences, measured [Formula: see text]-divergences, and sandwiched and [Formula: see text]-[Formula: see text]-Rényi divergences. In this paper, we give a systematic overview of the various concepts of quantum [Formula: see text]-divergences, with a main focus on their monotonicity under quantum operations, and the implications of the preservation of a quantum [Formula: see text]-divergence by a quantum operation. In particular, we compare the standard and the maximal [Formula: see text]-divergences regarding their ability to detect the reversibility of quantum operations. We also show that these two quantum [Formula: see text]-divergences are strictly different for non-commuting operators unless [Formula: see text] is a polynomial, and obtain some analogous partial results for the relation between the measured and the standard [Formula: see text]-divergences. We also study the monotonicity of the [Formula: see text]-[Formula: see text]-Rényi divergences under the special class of bistochastic maps that leave one of the arguments of the Rényi divergence invariant, and determine domains of the parameters [Formula: see text] where monotonicity holds, and where the preservation of the [Formula: see text]-[Formula: see text]-Rényi divergence implies the reversibility of the quantum operation.