Quantum Optimal Transport Barycenters: Existence, Duality, and Gaussian Rigidity

This paper addresses a complex problem in quantum information theory by developing a Quantum Optimal Transport (QOT) barycenter framework, analogous to classical Wasserstein barycenters, specifically for quantum states and quantum channels. The analysis is rigorous, employing advanced techniques from convex optimization and functional analysis.

Here is a detailed and comprehensive summary of the paper's main contributions, structure, key results, and technical machinery:

Comprehensive Research Summary: Quantum Optimal Transport Barycenters

This research introduces a novel Quantum Optimal Transport (QOT) barycenter framework designed to generalize classical Wasserstein barycenters to the quantum domain. The central goal is to establish existence and duality results for these quantum barycenters on separable Hilbert spaces, while providing a unified treatment across both the quantum-state formulation and the quantum-channel formulation.

I. Core Framework and Unified Theory

The paper constructs a variational theory for QOT barycenters, which serves as an analog to classical optimal transport problems. A key achievement is the unification of two distinct formulations:

1. Quantum-State Formulation: Utilizing canonical observables on separable Hilbert spaces (using Weyl systems).

2. Quantum-Channel Formulation: Addressing the problem in terms of quantum channels.

This unification is achieved by expressing both barycenter problems in a common, unified form involving Weyl systems:

B tau (alpha):= in S(L 2(R m)) X N s=1 alpha s QW 2 2, tau (, s) = in S(L 2(R m)) s in[N] pi s in tau (, s) X N s=1 alpha s Tr C(tau) pi s

where C(tau) is the cost operator, and tau represents the set of feasible couplings.

The framework is robust enough to handle a broad class of possibly unbounded transport costs, specifically considering two major classes:

1. Compact-resolvent cost operators.

2. Quadrature-generated costs, which are defined as noncommutative analogues of the quadratic Euclidean cost, based on canonical generators of regular Weyl systems.

II. Main Contributions and Key Results

The paper is organized around three primary pillars: Existence/Duality, Gaussian Reduction/Uniqueness, and Semiclassical Limits/Thermal Inputs.

1. Existence and Duality (Theorem 2.1)

The authors establish the foundational results by developing a variational theory that guarantees existence and Kantorovich duality for QOT barycenters under specific assumptions regarding cost operators (confinement and finite-cost feasibility). This provides a general mathematical foundation for the entire framework, covering both finite-dimensional systems and continuous-variable systems.

2. Gaussian Reduction and Global Uniqueness (Theorem 2.5 & Theorem 2.6)

This is arguably the most significant result concerning the behavior of quantum states:

• Gaussian Minimizer: For canonical quadratic costs, it is proven that Gaussian input states admit a Gaussian barycenter. Furthermore, this minimum is characterized by a finite-dimensional convex optimization problem over covariance matrices (Theorem 2.5).

• Global Rigidity Theorem: The central claim is the global rigidity theorem: If at least one input state among the set of inputs is faithful, then the resulting 2-QW barycenter is unique among all quantum states and must necessarily be Gaussian (Theorem 2.6).

• Faithfulness: This condition ensures uniqueness. It is sufficient but not necessary; families of pure inputs can still determine a unique barycenter, although partially pure nonfaithful Gaussian inputs may admit multiple barycenters (Example 5.31).

3. Semiclassical Limit and Thermal Inputs (Theorem 2.7 & Corollary 2.8)

The framework connects the quantum results to classical physics:

• Semiclassical Convergence: For Gaussian means and fixed positive-definite covariances independent of, Theorem 2.7 proves the semiclassical convergence of the quantum barycenter covariances to the covariance of the corresponding classical Gaussian 2-Wasserstein barycenter as to 0+. The associated Wigner distributions converge weakly to the classical Gaussian measure.

• Thermal Inputs: Corollary 2.8 provides an explicit characterization for a unique barycenter when inputs are displaced isotropic thermal Gaussian states, yielding an explicit covariance formula in the quantum-state formulation and a one-dimensional convex characterization in the quantum-channel formulation.

III. Technical Machinery and Proof Structure

The proof structure relies heavily on advanced functional analysis tools:

• Duality: The authors utilize Fenchel–Rockafellar duality to establish the relationship between the primal minimization problem and its dual counterpart (Proposition C.13). This duality is crucial for characterizing optimal couplings via a complementary-slackness relation.

• Uniqueness Proof Mechanism: Uniqueness of the full quantum state is established by showing that covariance uniqueness implies uniqueness of the Gaussian projection, which then, under the faithful-input assumption, forces the entire barycenter to be unique and Gaussian (Corollary 5.17). This relies on reconstructing an optimal coupling from its second marginal using a specific covariance slackness relation (Proposition 5.24) to show that this coupling is itself Gaussian (Corollary 5.22).

• Symplectic Geometry: The paper employs symplectic coordinates to analyze the covariance slackness relation, utilizing the unitary–orthosymplectic identification (Proposition D.2) to relate matrix groups (U(m) and Sp(2m, R) O(2m)).

• Quantum Channel Representation: A key lemma (Lemma D.4) details how a state can be represented as a quantum channel pi via a Williamson decomposition, linking the state formulation to the channel formulation.

IV. Conclusion and Significance

The paper successfully establishes a rigorous, unified mathematical structure for Quantum Optimal Transport barycenters. It moves beyond mere existence by providing deep structural results concerning Gaussian rigidity—a powerful statement about how quantum inputs constrain the resulting barycenter's form (forcing it to be Gaussian if an input is faithful). The framework bridges the gap between continuous-variable quantum systems and their classical limits, offering explicit formulas for thermal states.

The technical rigor, demonstrated through detailed proofs involving Fenchel conjugates, symplectic transformations, and specific cost operator analysis (e.g., Proposition E.1 on isotropic symmetrization), suggests a high degree of mathematical soundness suitable for publication in a top-tier journal. The identified nonuniqueness examples (Example 5.31) are also valuable as they precisely delineate the boundary conditions under which the rigidity theorem fails, providing necessary context for the main result.

1. Based on Theorem 2.6, AI systems can achieve unique quantum state representation for 2-QW barycenters when at least one input Gaussian state is faithful, as this forces the barycenter to be unique and Gaussian: if at least one Gaussian input is faithful, then the barycenter is unique among all quantum states and is necessarily Gaussian.

2. AI systems can perform semiclassical convergence to classical models by using Theorem 2.7, which proves that "for sufficiently small ħ > 0, let σˆ ħ s be the Gaussian state with mean ms and covariance Σs. Since Σs≻ 0, all these states are faithful for sufficiently small ħ. Hence Theorem 2.6 gives, for each τ ∈ ±1, a unique 2-QW barycenter ρˆ ħτ."

3. AI systems can solve complex quantum transport problems by employing the unified QOT framework (Equation 1.4), allowing them to treat both the quantum-state formulation and the quantum-channel formulation simultaneously through the parameter τ.

4. AI systems can optimize transport costs under finite energy constraints by utilizing the strong duality result (Theorem 2.2) and its dual formula (Equation 2.3), which provides a rigorous mathematical framework for minimizing the objective function subject to moment bounds: B C,α A,E (σ) = sup [X N s=1 αs Tr(Vsσs) − X l∈[k] ηlEl: ((Us, Vs)s∈[N], (ηl)l∈[k]) ∈ PotC,αA,E(σ)].

5. AI systems can identify the unique optimal covariance matrix for Gaussian barycenters by solving the finite-dimensional optimization problem (Equation 2.11), which reduces to min Σ∈CovQ m, As∈Corrτ (Tr(Σ) +X N s=1 αs Tr(Σs) − 2 Tr(As)).

6. AI systems can reconstruct the underlying quantum state from its first and second moments when a covariance slackness relation is satisfied by using the reconstruction principle: At fixed first moments, every state πˆ satisfying a covariance slackness relation with a positive semidefinite matrix Z can be written as πˆ = ΦG(ˆη), where the fixed Gaussian channel ΦG satisfies Ran(XΦ) = ker(ReZ).

7. AI systems can solve for displaced thermal inputs using Corollary 2.8, which provides an explicit characterization of the unique barycenter covariance: For τ = +1, Proposition 5.28 gives B+1(σˆ th α) = 2m min w≥ħ/2 [X N s=1 αs 'r w − ħ/2 - rws - ħ/2 squared + 2mħ + X N s=1 αs m − ms 2].

8. AI systems can determine the unique barycenter for displaced thermal inputs by finding the minimizer of the explicit characterization derived in Corollary 5.29: B+1(σˆ th α) = [X N s=1 (2mαs 'r w+1 − ħ/2 - rws - ħ/2 squared + αs m − ms 2) + 2mħ] = 2m (X N s=1 αs [ws − ħ/2 - rws - ħ/2 2]) + (X N s=1 αs ms squared − X N s=1 αsms) + 2mħ.

9. AI systems can analyze nonuniqueness in quantum transport by identifying the boundary where distinct barycenters share identical first and second moments, as shown in Example 5.31: "Example 5.31 exhibits two distinct sources of nonuniqueness: the covariance problem Pτ,cov from (2.11) itself has multiple optimizers, and even after fixing the optimal covariance Στ0, the first and second moments do not determine the barycenter as a quantum state."

We develop a Quantum Optimal Transport (QOT) barycenter framework for quantum states, as an analog of Wasserstein barycenters, and establish existence and duality results for a broad class of possibly unbounded transport costs on separable Hilbert spaces. Our framework provides, in particular, a unified treatment of 2-quantum Wasserstein (QW) barycenters in both the quantum-state and quantum-channel formulations by specializing to the canonical quadratic cost operators associated with the 2-quantum Wasserstein distances of Caglioti--Golse--Mouhot--Paul and De Palma--Trevisan. The central results concern Gaussian input states and, in particular, Gaussian rigidity: whether Gaussian input states force the 2-QW barycenter itself to be Gaussian and uniquely determined when the minimization is taken over all quantum states. We first show that the barycenter problem admits a Gaussian minimizer and reduces to a finite-dimensional convex optimization problem over covariance matrices. The main difficulty is that uniqueness of the optimal covariance does not, in general, imply uniqueness of the underlying quantum state. We bridge this gap through a state-reconstruction principle under covariance complementary slackness that upgrades uniqueness of the optimal covariance to uniqueness of the full quantum state, and thereby prove a global rigidity theorem: if at least one Gaussian input is faithful, then the barycenter is unique among all quantum states and is necessarily Gaussian. Faithfulness is sufficient but not necessary: families of pure inputs still determine a unique barycenter, whereas partially pure nonfaithful Gaussian inputs may admit multiple barycenters.