Wasserstein distances and divergences of order p by quantum channels
Pith reviewed 2026-05-23 05:22 UTC · model grok-4.3
The pith
Quantum channels realize p-order Wasserstein distances and divergences between quantum states.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
By minimizing a p-cost functional over quantum channels that send one density operator to another, one obtains well-defined p-Wasserstein distances and divergences. These objects are nonnegative, symmetric, and zero only on equal states. The triangle inequality holds for the quadratic case as soon as one of the three states is pure.
What carries the argument
The p-cost functional minimized over quantum channels that map one state to another.
If this is right
- The p-Wasserstein objects extend the quadratic case to arbitrary positive orders.
- Positivity, symmetry, and vanishing only on equal states hold by construction.
- The triangle inequality for quadratic divergences requires purity of only one state.
- Unitary invariance and other geometric features carry over from the channel model.
Where Pith is reading between the lines
- When all states commute the new objects are expected to recover the classical p-Wasserstein distances on probability vectors.
- The relaxation to a single pure state may enlarge the set of triples for which the inequality can be used in concrete calculations.
- The same channel-based minimization could be applied to other cost functions beyond powers of the distance.
Load-bearing premise
Quantum transport between states is realized exactly by quantum channels.
What would settle it
An explicit triple of mixed quantum states for which the quadratic Wasserstein divergence defined by channels violates the triangle inequality.
read the original abstract
We introduce a non-quadratic generalization of the quantum mechanical optimal transport problem introduced in [De Palma and Trevisan, Ann. Henri Poincar\'e, {\bf 22} (2021), 3199-3234] where quantum channels realize the transport. Relying on this general machinery, we introduce $p$-Wasserstein distances and divergences and study their fundamental geometric properties. Finally, we prove triangle inequality for quadratic Wasserstein divergences under the sole assumption that an arbitrary one of the states involved is pure, which is a generalization of our previous result in this direction.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript generalizes the quantum optimal transport problem of De Palma and Trevisan (2021) by replacing the quadratic cost with a p-order cost realized via quantum channels. It defines corresponding p-Wasserstein distances and divergences, examines their geometric properties (including positivity, symmetry, and contractivity), and proves the triangle inequality for the quadratic (p=2) case under the relaxed assumption that only one of the three states is pure.
Significance. If the derivations hold, the work supplies a coherent non-quadratic extension of channel-based quantum optimal transport and relaxes the purity hypothesis in the triangle inequality, which is a concrete technical advance over the authors' prior result. The construction introduces no free parameters and stays within the existing De Palma-Trevisan framework.
major comments (1)
- [Section 4] The triangle inequality (presumably Theorem 4.3 or equivalent) is stated to hold when exactly one state is pure. The proof sketch in the text appears to reduce the general case to the pure-state case via channel properties, but the reduction step that removes the purity requirement on the other two states is not spelled out in sufficient detail to verify the bound on the cross terms.
minor comments (2)
- [Section 2] Notation for the p-cost functional is introduced without an explicit comparison table to the quadratic case; adding such a table would clarify continuity at p=2.
- [Introduction] The abstract claims 'we prove triangle inequality … which is a generalization of our previous result'; the introduction should cite the precise prior theorem being generalized.
Simulated Author's Rebuttal
We thank the referee for their thorough review and constructive feedback. We are pleased that the significance of our work is recognized. Below we address the major comment point by point.
read point-by-point responses
-
Referee: [Section 4] The triangle inequality (presumably Theorem 4.3 or equivalent) is stated to hold when exactly one state is pure. The proof sketch in the text appears to reduce the general case to the pure-state case via channel properties, but the reduction step that removes the purity requirement on the other two states is not spelled out in sufficient detail to verify the bound on the cross terms.
Authors: We thank the referee for highlighting this point. Upon re-examination, we agree that the reduction step in the proof of the triangle inequality could benefit from additional detail to clarify how the purity of one state allows bounding the cross terms via channel properties. In the revised version of the manuscript, we will expand this part of the proof, providing explicit steps for the reduction and the estimation of the relevant terms. revision: yes
Circularity Check
Minor self-citation of prior result; central derivation remains independent
full rationale
The manuscript extends the quantum channel-based optimal transport framework introduced in the external reference De Palma and Trevisan (2021) to define p-Wasserstein distances and divergences, then supplies a new proof of the triangle inequality for the quadratic case under the relaxed assumption that only one of three states is pure. This is explicitly described as a generalization of the authors' own earlier result, constituting a minor self-citation that is not load-bearing because the present paper provides the required extensions and modified proof steps. No step reduces a claimed prediction or uniqueness statement to a fitted parameter, self-definition, or unverified self-citation chain; the derivation chain is self-contained against the cited external foundation.
Axiom & Free-Parameter Ledger
axioms (1)
- domain assumption Quantum channels realize the transport maps between states
Forward citations
Cited by 3 Pith papers
-
Strong Kantorovich duality for quantum optimal transport with generic cost and optimal couplings on quantum bits
Establishes Kantorovich duality for linearized non-quadratic quantum optimal transport realized by channels, determines optimal primal-dual solutions for qubits under state restrictions, and proves the triangle inequa...
-
Relations between different definitions of the quantum Wasserstein distance for qubits
Two quantum Wasserstein distance definitions coincide for qubits with single-operator cost functions, implying the self-distance equals the Wigner-Yanase skew information.
-
Quantum Wasserstein distance and its relation to several types of fidelities
Proves equalities among quantum Wasserstein distances obtained from optimizations over general versus separable bipartite states and shows relations to Uhlmann-Jozsa fidelity and superfidelity, including equality for qubits.
Reference graph
Works this paper leans on
-
[1]
Order p quantum Wasserstein distances from couplings
Emily Beatty and Daniel Stilck Fran¸ ca. Order p quantum Wasserstein distances from couplings. 2024. arXiv:2402.16477
-
[2]
A free probability an alogue of the Wasserstein metric on the trace-state space
Philippe Biane and Dan Voiculescu. A free probability an alogue of the Wasserstein metric on the trace-state space. Geom. Funct. Anal., 11(6):1125–1138, 2001. doi:10.1007/s00039-001-8226-4
-
[3]
Monotonicity of the quantum 2-Wasserstein dis tance, 2022
Rafa/suppress l Bistro´ n, Micha/suppress l Eckstein, and Karol˙Zyczkowski. Monotonicity of the quantum 2-Wasserstein dis tance, 2022. arXiv:2204.07405
-
[4]
Distribution’s template estimate with W asserstein m et- rics
Emmanuel Boissard, Thibaut Le Gouic, and Jean-Michel Lo ubes. Distribution’s template estimate with W asserstein m et- rics. Bernoulli, 21(2):740–759, 2015. doi:10.3150/13-BEJ585
-
[5]
Metric property of quantum wasserstein diver gences
Gergely Bunth, J´ ozsef Pitrik, Tam´ as Titkos, and D´ aniel Virosztek. Metric property of quantum wasserstein diver gences. Phys. Rev. A , 110:022211, Aug 2024. doi:10.1103/PhysRevA.110.022211
-
[6]
Subgeometric rates of convergence of Ma rkov processes in the Wasserstein metric
Oleg Butkovsky. Subgeometric rates of convergence of Ma rkov processes in the Wasserstein metric. Ann. Appl. Probab. , 24(2):526–552, 2014. doi:10.1214/13-AAP922
-
[7]
Q uantum optimal transport is cheaper
Emanuele Caglioti, Fran¸ cois Golse, and Thierry Paul. Q uantum optimal transport is cheaper. J. Stat. Phys. , 181(1):149– 162, 2020. doi:10.1007/s10955-020-02571-7
-
[8]
T owards optimal transport for quantum densities, 2021
Emanuele Caglioti, Fran¸ cois Golse, and Thierry Paul. T owards optimal transport for quantum densities, 2021. arXiv:2101.03256
-
[9]
Eric Carlen. Dynamics and quantum optimal transport: th ree lectures on quantum entropy and quantum Markov semigroups. In Optimal transport on quantum structures , pages 29–89. Springer, Cham, 2024. URL: https://doi.org/10.1007/978-3-031-50466-2_2 , doi:10.1007/978-3-031-50466-2\_2 . 24 GERGELY BUNTH, J ´OZSEF PITRIK, TAM ´AS TITKOS, AND D ´ANIEL VIROSZTEK
-
[10]
Eric A. Carlen and Jan Maas. An analog of the 2-Wasserste in metric in non-commutative probability under which the fermionic Fokker-Planck equation is gradient flow for th e entropy. Comm. Math. Phys. , 331(3):887–926, 2014. doi:10.1007/s00220-014-2124-8
-
[11]
Eric A. Carlen and Jan Maas. Gradient flow and entropy ine qualities for quantum Markov semigroups with detailed bala nce. J. Funct. Anal. , 273(5):1810–1869, 2017. doi:10.1016/j.jfa.2017.05.003
-
[12]
Eric A. Carlen and Jan Maas. Non-commutative calculus, optimal transport and functional inequalities in dissipat ive quantum systems. J. Stat. Phys. , 178(2):319–378, 2020. doi:10.1007/s10955-019-02434-w
-
[13]
Concentration of q uantum states from quantum functional and transportation c ost inequalities
Nilanjana Datta and Cambyse Rouz´ e. Concentration of q uantum states from quantum functional and transportation c ost inequalities. J. Math. Phys. , 60(1):012202, 22, 2019. doi:10.1063/1.5023210
-
[14]
Relating relative entropy, optimal transport and Fisher information: a quant um HWI inequality
Nilanjana Datta and Cambyse Rouz´ e. Relating relative entropy, optimal transport and Fisher information: a quant um HWI inequality. Ann. Henri Poincar´ e, 21(7):2115–2150, 2020. doi:10.1007/s00023-020-00891-8
-
[15]
Personal communic ation at the QOTTA 2024 Workshop in Cortona, in September 2024
Giacomo De Palma and Dario Trevisan. Personal communic ation at the QOTTA 2024 Workshop in Cortona, in September 2024
work page 2024
-
[16]
Quantum optimal tr ansport with quantum channels
Giacomo De Palma and Dario Trevisan. Quantum optimal tr ansport with quantum channels. Ann. Henri Poincar´ e, 22(10):3199–3234, 2021. doi:10.1007/s00023-021-01042-3
-
[17]
Quantum optimal tr ansport: quantum channels and qubits
Giacomo De Palma and Dario Trevisan. Quantum optimal tr ansport: quantum channels and qubits. In Jan Maas, Simone Rademacher, Tam´ as Titkos, and D´ aniel Virosztek, editors, Optimal Transport on Quantum Structures . Springer Nature,
-
[18]
Geometry of cuts and metrics , volume 15 of Algorithms and Combinatorics
Michel Marie Deza and Monique Laurent. Geometry of cuts and metrics , volume 15 of Algorithms and Combinatorics . Springer, Heidelberg, 2010. First softcover printing of th e 1997 original [MR1460488]. doi:10.1007/978-3-642-04295-9
-
[19]
Rocco Duvenhage. Optimal quantum channels. Phys. Rev. A , 104(3):Paper No. 032604, 8, 2021. doi:10.1103/physreva.104.032604
-
[20]
Quadratic Wasserstein metrics for vo n Neumann algebras via transport plans
Rocco Duvenhage. Quadratic Wasserstein metrics for vo n Neumann algebras via transport plans. J. Operator Theory , 88(2):289–308, 2022. arXiv:2012.03564
-
[21]
E xtending quantum detailed balance through optimal transpo rt
Rocco Duvenhage, Samuel Skosana, and Machiel Snyman. E xtending quantum detailed balance through optimal transpo rt. Reviews in Mathematical Physics , 2024. arXiv:2206.15287, doi:10.1142/S0129055X24500405
-
[22]
A. Figalli and F. Maggi. On the shape of liquid drops and c rystals in the small mass regime. Arch. Ration. Mech. Anal. , 201(1):143–207, 2011. doi:10.1007/s00205-010-0383-x
-
[23]
A. Figalli, F. Maggi, and A. Pratelli. A mass transporta tion approach to quantitative isoperimetric inequalities . Invent. Math., 182(1):167–211, 2010. doi:10.1007/s00222-010-0261-z
-
[24]
Quantum Monge-Kantorovich Prob- lem and Transport Distance between Density Matrices
Shmuel Friedland, Micha/suppress lEckstein, Sam Cole, and Karo l ˙Zyczkowski. Quantum Monge-Kantorovich Prob- lem and Transport Distance between Density Matrices. Phys. Rev. Lett. , 129(11):Paper No. 110402, 2022. doi:10.1103/physrevlett.129.110402
-
[25]
Optimal Transport Methods in Economics
Alfred Galichon. Optimal Transport Methods in Economics . Princeton University Press, 1 edition, 2016. URL: https://EconPapers.repec.org/RePEc:pup:pbooks:10870
work page 2016
-
[26]
Onthe mean field and classical limits of quantum mechanics
Fran¸ cois Golse, Cl´ ement Mouhot, and Thierry Paul. Onthe mean field and classical limits of quantum mechanics. Comm. Math. Phys. , 343(1):165–205, 2016. doi:10.1007/s00220-015-2485-7
-
[27]
The Schr¨ odinger equation in the mean-field and semiclassical regime
Fran¸ cois Golse and Thierry Paul. The Schr¨ odinger equation in the mean-field and semiclassical regime. Arch. Ration. Mech. Anal., 223(1):57–94, 2017. doi:10.1007/s00205-016-1031-x
-
[28]
W ave packets and the q uadratic Monge-Kantorovich distance in quantum mechanics
Fran¸ cois Golse and Thierry Paul. W ave packets and the q uadratic Monge-Kantorovich distance in quantum mechanics . C. R. Math. Acad. Sci. Paris , 356(2):177–197, 2018. doi:10.1016/j.crma.2017.12.007
-
[29]
Empirical measures and quantum mechanics: applications to the mean-field limit
Fran¸ cois Golse and Thierry Paul. Empirical measures and quantum mechanics: applications to the mean-field limit. Comm. Math. Phys. , 369(3):1021–1053, 2019. doi:10.1007/s00220-019-03357-z
-
[30]
Mean-field and classi cal limit for the N -body quantum dynamics with Coulomb interac- tion
Fran¸ cois Golse and Thierry Paul. Mean-field and classi cal limit for the N -body quantum dynamics with Coulomb interac- tion. Comm. Pure Appl. Math. , 75(6):1332–1376, 2022. doi:10.1002/cpa.21986
-
[31]
Optimal transport ps eudometrics for quantum and classical densities
Fran¸ cois Golse and Thierry Paul. Optimal transport ps eudometrics for quantum and classical densities. J. Funct. Anal. , 282(9):Paper No. 109417, 53, 2022. doi:10.1016/j.jfa.2022.109417
-
[32]
Fran¸ cois Golse and Thierry Paul. Quantitative observ ability for the Schr¨ odinger and Heisenberg equations: an o ptimal transport approach. Math. Models Methods Appl. Sci. , 32(5):941–963, 2022. doi:10.1142/S021820252250021X
-
[33]
Martin Hairer and Jonathan C. Mattingly. Spectral gaps in Wasserstein distances and the 2D stochastic Navier-Stok es equations. Ann. Probab., 36(6):2050–2091, 2008. doi:10.1214/08-AOP392
-
[34]
Martin Hairer, Jonathan C. Mattingly, and Michael Sche utzow. Asymptotic coupling and a general form of Har- ris’ theorem with applications to stochastic delay equatio ns. Probab. Theory Related Fields , 149(1-2):223–259, 2011. doi:10.1007/s00440-009-0250-6
-
[35]
Learning elastic costs to shape monge displacements, 2024
Michal Klein, Aram-Alexandre Pooladian, Pierre Ablin , Eug` ene Ndiaye, Jonathan Niles-W eed, and Marco Cuturi. Learning elastic costs to shape monge displacements, 2024. URL: https://arxiv.org/abs/2306.11895, arXiv:2306.11895. p-W ASSERSTEIN DISTANCES AND DIVERGENCES BY QUANTUM CHANNEL S 25
-
[36]
Soheil Kolouri, Se Rim Park, and Gustavo K. Rohde. The ra don cumulative distribution transform and its application to image classification. IEEE Trans. Image Process. , 25(2):920–934, 2016. doi:10.1109/TIP.2015.2509419
-
[37]
Soheil Kolouri, Akif B. Tosun, John A. Ozolek, and Gusta vo K. Rohde. A continuous linear optimal transport approach for pattern analysis in image datasets. Pattern Recognition , 51:453–462, 2016. URL: https://www.sciencedirect.com/science/article/pii/S0031320315003507, doi:10.1016/j.patcog.2015.09.019
-
[38]
W asserstein task embedding for measuring task similarities
Xinran Liu, Yikun Bai, Yuzhe Lu, Andrea Soltoggio, and S oheil Kolouri. W asserstein task embedding for measuring task similarities. Neural Networks , 181:106796, 2025. URL: https://www.sciencedirect.com/science/article/pii/S0893608024007202, doi:10.1016/j.neunet.2024.106796
-
[39]
John Lott and C´ edric Villani. Ricci curvature for metr ic-measure spaces via optimal transport. Ann. of Math. (2) , 169(3):903–991, 2009. doi:10.4007/annals.2009.169.903
-
[40]
Optimal transport on quantum structures
Jan Maas, Simone Rademacher, Tam´ as Titkos, and D´ anie l Virosztek, editors. Optimal transport on quantum structures . Springer, Cham, 2024. doi:10.1007/978-3-031-50466-2
-
[41]
Se Rim Park, Soheil Kolouri, Shinjini Kundu, and Gustav o K. Rohde. The cumulative distribution trans- form and linear pattern classification. Applied and Computational Harmonic Analysis , 45(3):616–641, 2018. URL: https://www.sciencedirect.com/science/article/pii/S1063520317300076, doi:10.1016/j.acha.2017.02.002
-
[42]
Computational Optimal Transport: With Ap- plications to Data Science
Gabriel Peyr´ e and Marco Cuturi. Computational optima l transport: With applications to data science. Foun- dations and Trends ® in Machine Learning , 11(5-6):355–607, 2019. URL: http://dx.doi.org/10.1561/2200000073, doi:10.1561/2200000073
-
[43]
On Wasserstein two-sample testing and related famili es of nonparametric tests
Aaditya Ramdas, Nicol´ as Garc ´ ıa Trillos, and Marco Cu turi. On Wasserstein two-sample testing and related famili es of nonparametric tests. Entropy, 19(2):Paper No. 47, 15, 2017. doi:10.3390/e19020047
-
[44]
Geoffrey Schiebinger, Jian Shu, Marcin Tabaka, Brian Cl eary, Vidya Subramanian, Aryeh Solomon, Joshua Gould, Siyan Liu, Stacie Lin, Peter Berube, Lia Lee, Jenny Ch en, Justin Brumbaugh, Philippe Rigollet, Konrad Hochedlinger, Rudolf Jaenisch, Aviv Regev, and Eric S. Lander. Optimal-transport analysis of single- cell gene expression identifies developmental...
-
[45]
Sanvesh Srivastava, Cheng Li, and David B. Dunson. Scal able bayes via barycenter in wasserstein space. J. Mach. Learn. Res., 19(1):312–346, January 2018
work page 2018
-
[46]
Karl-Theodor Sturm. On the geometry of metric measure s paces. I. Acta Math. , 196(1):65–131, 2006. doi:10.1007/s11511-006-0002-8
-
[47]
Karl-Theodor Sturm. On the geometry of metric measure s paces. II. Acta Math. , 196(1):133–177, 2006. doi:10.1007/s11511-006-0003-7
-
[48]
Quantum Wasserstein distance based on an optimization over separable states
G´ eza T´ oth and J´ ozsef Pitrik. Quantum Wasserstein distance based on an optimization over separable states. Quantum, 7:1143, October 2023. doi:10.22331/q-2023-10-16-1143
-
[49]
C´ edric Villani. Optimal transport, volume 338 of Grundlehren der mathematischen Wissenschaften [Fundamen tal Princi- ples of Mathematical Sciences] . Springer-Verlag, Berlin, 2009. Old and new. doi:10.1007/978-3-540-71050-9
-
[50]
von Renesse and Karl-Theodor Sturm
Max-K. von Renesse and Karl-Theodor Sturm. Transport i nequalities, gradient estimates, entropy, and Ricci curva ture. Comm. Pure Appl. Math. , 58(7):923–940, 2005. doi:10.1002/cpa.20060
-
[51]
W ei W ang, Dejan Slepˇ cev, Saurav Basu, John A. Ozolek, a nd Gustavo K. Rohde. A linear optimal transportation framework for quantifying and visualizing variations in se ts of images. Int. J. Comput. Vis. , 101(2):254–269, 2013. doi:10.1007/s11263-012-0566-z
-
[52]
A dual formula for the noncommutative t ransport distance
Melchior Wirth. A dual formula for the noncommutative t ransport distance. J. Stat. Phys. , 187(2):Paper No. 19, 18, 2022. doi:10.1007/s10955-022-02911-9
-
[53]
The Monge metric onthe sphere and geometry of quantum states
Karol ˙Zyczkowski and W ojciech S/suppress lomczy´ nski. The Monge metric onthe sphere and geometry of quantum states. J. Phys. A, 34(34):6689–6722, 2001. doi:10.1088/0305-4470/34/34/311
-
[54]
The Monge distance between quantum states
Karol ˙Zyczkowski and W ojeciech S/suppress lomczy´ nski. The Monge distance between quantum states. J. Phys. A , 31(45):9095–9104,
-
[55]
doi:10.1088/0305-4470/31/45/009
-
[56]
Saurav Basu and Soheil Kolouri and Gustavo K. Rohde. Det ecting and visualizing cell phenotype dif- ferences from microscopy images using transport-based mor phometry. Proceedings of the National Acad- emy of Sciences , 111(9):3448–3453, 2014. URL: https://www.pnas.org/doi/abs/10.1073/pnas.1319779111, arXiv:https://www.pnas.org/doi/pdf/10.1073/pnas.131977...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.