Quantum $f$-divergences and Their Local Behaviour: An Analysis via Relative Expansion Coefficients

• URL: http://arxiv.org/abs/2510.06183v2
• Date: Tue, 14 Oct 2025 16:40:38 GMT
• Title: Quantum $f$-divergences and Their Local Behaviour: An Analysis via Relative Expansion Coefficients
• Authors: Shreyas Iyer, Peixue Wu, Paula Belzig, Graeme Smith,
• Abstract summary: We study contraction and expansion coefficients, which can be combined into a single emphrelative expansion coefficient<n>We identify new families of $f$ for which the global ($f$ divergence) and local (Riemannian) relative expansion coefficients coincide for every pair of channels.<n>We prove a reverse quantum Markov convergence theorem, converting positive expansion coefficients into quantitative lower bounds on the convergence rate.
• Score: 4.30484058393522
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Any reasonable measure of distinguishability of quantum states must satisfy a data processing inequality, that is, it must not increase under the action of a quantum channel. We can ask about the proportion of information lost or preserved and this leads us to study contraction and expansion coefficients respectively, which can be combined into a single \emph{relative expansion coefficient}. We focus on two prominent families: (i) standard quantum $f$ divergences and (ii) their local (second-order) behaviour, which induces a monotone Riemannian semi-norm (that is linked to the $\chi^2$ divergence). Building on prior work, we identify new families of $f$ for which the global ($f$ divergence) and local (Riemannian) relative expansion coefficients coincide for every pair of channels, and we clarify how exceptional such exact coincidences are. Beyond equality, we introduce an \emph{equivalence} framework that transfers qualitative properties such as strict positivity uniformly across different relative expansion coefficients. Leveraging the link between equality in the data processing inequality (DPI) and channel reversibility, we apply our framework of relative expansion coefficients to approximate recoverability of quantum information. Using our relative expansion results for primitive channels, we prove a reverse quantum Markov convergence theorem, converting positive expansion coefficients into quantitative lower bounds on the convergence rate.

Related papers

• From Tail Universality to Bernstein-von Mises: A Unified Statistical Theory of Semi-Implicit Variational Inference [0.12183405753834557]
  Semi-implicit variational inference (SIVI) constructs approximate posteriors of the form $q() = int k(| z) r(dz)$<n>This paper develops a unified "approximation-optimization-statistics'' theory for such families.
  arXiv  Detail & Related papers  (2025-12-05T19:26:25Z)
• A Unified Approach to Quantum Contraction and Correlation Coefficients [10.128808054306187]
  We introduce a family of non-commutative $L2(p)$ spaces induced by operator monotone functions.<n>We identify families of quantum maximal correlation coefficients and the quantum $chi2$-divergences.
  arXiv  Detail & Related papers  (2025-05-21T08:58:45Z)
• Generalization Bounds for Quantum Learning via Rényi Divergences [45.45698347373077]
  This work advances the theoretical understanding of quantum learning by establishing a new family of upper bounds on the expected generalization error of quantum learning algorithms.<n>Our primary contribution is the derivation of these bounds in terms of quantum and classical R'enyi divergences.
  arXiv  Detail & Related papers  (2025-05-16T09:21:31Z)
• Evolution of Gaussians in the Hellinger-Kantorovich-Boltzmann gradient flow [44.500314182711236]
  We derive explicit ordinary equations govern the evolution of mean, covariance, and mass under HK-Boltzmann flow.<n>We demonstrate exponential convergence to equilibrium through constructing a gradient structure associated with a Polyak-Lojasiewicz-type inequalities.<n>We extend our analysis to non-Gaussian targets exhibiting strong log-lambda-concavity.
  arXiv  Detail & Related papers  (2025-04-29T03:54:56Z)
• Quantum Doeblin Coefficients: Interpretations and Applications [7.432224771219169]
  In classical information theory, the Doeblin coefficient of a classical channel provides an efficiently computable upper bound on the total-variation contraction coefficient of the channel.<n>Here, we investigate quantum Doeblin coefficients as a generalization of the classical concept.<n>We outline various applications of quantum Doeblin coefficients, ranging from limitations on quantum machine learning algorithms to error mitigation protocols.
  arXiv  Detail & Related papers  (2025-03-28T18:45:44Z)
• A semiconcavity approach to stability of entropic plans and exponential convergence of Sinkhorn's algorithm [3.686530147760242]
  We study stability of bounds and convergence of Sinkhorn's algorithm for the entropic optimal transport problem.<n>New applications include subspace elastic costs, weakly log-concave marginals, marginals with light tails.
  arXiv  Detail & Related papers  (2024-12-12T12:45:31Z)
• A New Theoretical Perspective on Data Heterogeneity in Federated Optimization [39.75009345804017]
  In federated learning (FL), data heterogeneity is the main reason that existing theoretical analyses are pessimistic about the convergence rate. In particular, for many FL algorithms, the convergence rate grows dramatically when the number of local updates becomes large. This paper aims to bridge this gap between theoretical understanding and practical performance by providing a theoretical analysis from a new perspective.
  arXiv  Detail & Related papers  (2024-07-22T11:52:58Z)
• Improved Quantum Algorithms for Fidelity Estimation [77.34726150561087]
  We develop new and efficient quantum algorithms for fidelity estimation with provable performance guarantees. Our algorithms use advanced quantum linear algebra techniques, such as the quantum singular value transformation. We prove that fidelity estimation to any non-trivial constant additive accuracy is hard in general.
  arXiv  Detail & Related papers  (2022-03-30T02:02:16Z)
• Tight Exponential Analysis for Smoothing the Max-Relative Entropy and for Quantum Privacy Amplification [56.61325554836984]
  The max-relative entropy together with its smoothed version is a basic tool in quantum information theory. We derive the exact exponent for the decay of the small modification of the quantum state in smoothing the max-relative entropy based on purified distance.
  arXiv  Detail & Related papers  (2021-11-01T16:35:41Z)
• R\'enyi divergence inequalities via interpolation, with applications to generalised entropic uncertainty relations [91.3755431537592]
  We investigate quantum R'enyi entropic quantities, specifically those derived from'sandwiched' divergence. We present R'enyi mutual information decomposition rules, a new approach to the R'enyi conditional entropy tripartite chain rules and a more general bipartite comparison.
  arXiv  Detail & Related papers  (2021-06-19T04:06:23Z)
• On contraction coefficients, partial orders and approximation of capacities for quantum channels [2.9005223064604073]
  We revisit the notion of contraction coefficients of quantum channels, which provide sharper and specialized versions of the data processing inequality. A concept closely related to data processing is partial orders on quantum channels.
  arXiv  Detail & Related papers  (2020-11-11T18:11:15Z)
• Recoverability for optimized quantum $f$-divergences [22.04181157631236]
  We show that the absolute difference between the optimized $f$-divergence and its channel-processed version is an upper bound on how well one can recover a quantum state. Not only do these results lead to physically meaningful refinements of the data-processing inequality for the sandwiched R'enyi relative entropy, but they also have implications for perfect reversibility.
  arXiv  Detail & Related papers  (2020-08-04T15:56:12Z)
• Linear Last-iterate Convergence in Constrained Saddle-point Optimization [48.44657553192801]
  We significantly expand the understanding of last-rate uniqueness for Optimistic Gradient Descent Ascent (OGDA) and Optimistic Multiplicative Weights Update (OMWU) We show that when the equilibrium is unique, linear lastiterate convergence is achieved with a learning rate whose value is set to a universal constant. We show that bilinear games over any polytope satisfy this condition and OGDA converges exponentially fast even without the unique equilibrium assumption.
  arXiv  Detail & Related papers  (2020-06-16T20:53:04Z)
• Optimal Bounds between $f$-Divergences and Integral Probability Metrics [8.401473551081748]
  Families of $f$-divergences and Integral Probability Metrics are widely used to quantify similarity between probability distributions. We systematically study the relationship between these two families from the perspective of convex duality. We obtain new bounds while also recovering in a unified manner well-known results, such as Hoeffding's lemma.
  arXiv  Detail & Related papers  (2020-06-10T17:39:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.