Quantum Hamming metrics

Marc A. Rieffel · 2025

Read Quantum Hamming metrics by Marc A. Rieffel (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-23046.txt.

Introduction

Marc Rieffel develops in 2025 a quantum version of the Hamming distance, the simple count of positions where two words of the same length differ. A quantum form of the related transport distance between states was introduced a few years earlier by De Palma and co-authors.

In the quantum setting, letters give way to matrix algebras and words to stacks of them side by side. Rieffel works at the level of observables rather than states, building the seminorm that plays the role of the Hamming metric itself. Such norms act like rulers for quantum systems and set limits on how fast their states can change.

Catalog record

CREATORMarc A. Rieffel
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE77 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-23046

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Quantum Hamming metrics Marc A. Rieffel Abstract . Given the set of words of a given length for a given alphabet, the Hamming metric between two such words is the number of positions where the two words differ. A quantum version of the corresponding Kantorovich-Wasserstein metric on states was introduced in 2021 by De Palma, Marvian, Trevisan and Lloyd. For the quantum version the alphabet is replaced by a full matrix algebra, and the set of words is replaced by the tensor product of a corresponding number of copies of that full matrix algebra. While De Palma et al. work primarily at the level of states, they do obtain the corresponding seminorm (the quantum Hamming metric) on the algebra of observables that plays the role of assigning Lipschitz constants to functions. A suitable such seminorm on a unital C*-algebra is the current common method for defining a quantum metric on a C*-algebra. In this paper we will reverse the process, by first expressing the Hamming metric in terms of the C*-algebra of functions on the set of words, and then dropping the requirement that the algebra be commutative so as to obtain the quantum Hamming metric. From that we obtain the corresponding Kantorovich-Wasserstein metric on states. Along the way we show that many of the steps can be put in more general forms of some interest, notably for infinite-dimensional C*-algebras. 2020 Mathematics Subject Classification Primary 58B34; Secondary 81P45 46L05 1. Introduction In 2021 De Palma et al. published a paper [ 7 ] in which they construct quantum versions of the Kantorovich-Wasserstein metrics on probability measures corresponding to Hamming metrics. They present many basic properties of these quantum versions, and indicate strong connections with a variety of important topics in quantum information theory. Their paper has spawned a large number of papers that apply the ideas in their paper in many directions. See [ 3 , 6 , 8 , 9 , 16 , 18 , 21 , 24 , 30 , 31 , 32 , 36 , 15 ] and the references they contain. In the course of their paper the authors obtain a quantum version of the Hamming metrics themselves. In the present paper we reverse the process, and start with Hamming metrics themselves, and provide a natural path that leads to the definition of quantum Hamming metrics. With this definition in hand, one is led quite naturally to the corresponding quantum Kantorovich-Wasserstein metrics on the state spaces of the underlying quantum systems, which in [ 7 ] are taken as the definition of the quantum Kantorovich-Wasserstein metrics (and which they call “The quantum Wasserstein distance of order 1”). In the present paper we put various of the steps into suitable more general contexts, which provide more perspective for the steps. In particular, we show that many (but not all) of the steps work well for arbitrary, possibly infinite-dimensional, unital C*-algebras. We also show how the quantum Hamming metrics can be expressed in terms of Dirac operator type structures. Here is a quick outline of our approach. We note first that the Hamming metric comes from first putting the complete-graph metric on each alphabet in use, and then using the corresponding sum-metric on the cartesian product of the alphabets. We then show how to express these two steps in terms of the corresponding algebras of functions (the algebras of “observables”), so that the Hamming metric is expressed entirely in terms of these algebras. Since this reformulation does not mention points of spaces, nor use the commutativity of the algebras, we can replace the algebras with ones that are not commutative, such as full matrix algebras, to obtain our quantum version of the Hamming metrics. The metric data is then given by a seminorm that plays the role of attaching the Lipschitz constants to functions on a metric space. We also show how to express the metric data in terms of “Dirac operators”, from which a few additional properties become evident. From the formulas for the quantum Hamming metrics we then obtain the formulas for the corresponding Kantorovich-Wasserstein metrics on states, that in [ 7 ] are taken as the definition of the quantum Wasserstein-1 metrics on states. Necessarily many of the results in the present paper are counter-parts, perhaps with some generalization, of results in [ 7 ] . Kuperberg and Weaver [ 19 ] have introduced a different definition of quantum metric spaces, motivated by quantum error correction, and formulated in the setting of von Neumann algebras. They call their metric structure a “W*-metric”. We sketch here what it looks like just in the finite-dimensional case, which is sufficient for their quantum Hamming metric, for which see [ 19 ] and example 11.30 in [ 34 ] . For a finite-dimensional Hilbert space {\mathcal{H}} and a C*-algebra {\mathcal{A}} of operators on it, they define a W*-metric on {\mathcal{A}} to be a filtration V_{t} of operator systems (i.e. self-adjoint subspaces containing the identity operator) in {\mathcal{B}}({\mathcal{H}}) for t ranging over a finite set of non-negative numbers. The filtration is required to satisfy the properties that 1) V_{0} is the commutant of {\mathcal{A}} in {\mathcal{B}}({\mathcal{H}}) , 2) V_{s}V_{t}\subseteq V_{s+t} . The intuition is that V_{t} consists of operators that displace mass by at most distance t (or introduce at most t errors). The term “finite-propagation operators” is sometimes used. The authors show that their W*-metrics can sometimes lead to C*-metrics as used in our present paper, and that Dirac-type constructions lead to W*-metrics. But very little is said about states, so the W*-metric approach appears to go in a direction quite different than that of the paper [ 7 ] of De Palma et al. that we study here. In [ 7 ] the authors present many important properties and applications of their Wasserstein-1 distance on states, which, as indicated above, have been widely explored and extended in other papers both by the authors and by many other researchers. We will not discuss any of this in the present paper, but there is much that well warrants exploration in the style of the present paper. 2. Hamming metrics In this section we begin the discussion of ordinary Hamming metrics, aiming to put them in a form suitable for suggesting the quantum versions that we will discuss later. The setting is as follows. We are given a finite set with n elements, I , whose elements we use as indices. For each i\in I we are given a finite set, X_{i} , called the alphabet at position i . (For the quantum Hamming metric, for each i our alphabet can be a quantum finite set, that is, a full matrix algebra {\mathcal{A}}_{i}=M_{m_{i}}({\mathbb{C}}) with its structure as a C*-algebra.) Often it is assumed that all these alphabets are the same set, but we do not need to make this assumption. We then consider “words” of length n with the i th entry taken from X_{i} for each i . The set of these words can be identified with the product set X=\Pi_{i}X_{i} . (In the quantum setting there are no words, but since C(X)=\otimes C(X_{i}) , in the quantum setting we will use the C*-algebra \otimes{\mathcal{A}}_{i} .) The Hamming metric on

Read the complete text