Geometric Construction of Optimal Teleportation Witnesses

arXiv:2605.22226 · quant-ph · Submitted 2026-05-21 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: We're looking at the paper: "Geometric Construction of Optimal Teleportation Witnesses".

Mira: Here's the gist of the paper. The gist The geometric construction of optimal teleportation witnesses provides an operational necessary and sufficient criterion for identifying the teleportation usefulness of arbitrary two-qudit entangled states by solving a shortest distance problem to the set of useless states.

Kai: Geometric Construction Method The paper proposes a geometric method to construct optimal teleportation witnesses from the projection geometry at the unique closest state σ∗ ∈ S to a given state ρ, where S is the convex set of states useless for teleportation and D(ρ) itself serves as a necessary and sufficient condition for usefulness.

Mira: The construction involves finding the projection point σ∗ ∈ S by solving the shortest distance optimization problem between ρ and S.

Kai: This problem is formulated as min σ ρ − σF s.t.

Paper summary: Kai: So, we're looking at this paper today called Geometric Construction of Optimal Teleportation Witnesses. Basically, it proposes a way to build witnesses that tell us exactly when a two-qudit entangled state is useful for quantum teleportation.

Mira: Right, what does that mean for the actual usefulness of these states? They claim this geometric method gives us an operational necessary and sufficient criterion instead of just using some inequalities we've seen before.

Lev: So it’s not just saying a state is useless or useful based on a simple test; they are constructing something specific, like an optimal teleportation witness, from the geometry. That sounds like it needs to actually be runnable on hardware.

Kai: Exactly. The core idea is solving a shortest distance problem from your target state rho to this set S of states that are useless for teleportation. They find this closest point sigma star in S, and then they build the witness using the geometry around that point.

Mira: And they claim that the resulting shortest distance, D(rho), also acts as a necessary and sufficient condition for usefulness. That means if D(rho) is greater than zero, you know it works for teleportation.

Lev: If we were trying to implement this on superconducting qubits, finding that closest point sigma star sounds like a tough optimization problem because the set S is defined by continuous constraints from all possible local unitaries U(d). How do they actually solve that?

Kai: They tackle it with a two-layer iterative cutting-plane algorithm. The outer layer tries to find a candidate state, and the inner layer figures out which unitary constraint is being violated the most to refine the search space.

Mira: That systematic alternation until convergence sounds like it could work for finding that closest point sigma star, even if it’s computationally intensive. They say this procedure empirically converges to the optimal solution.

Lev: So, what does that convergence mean for us on real hardware? If we had to run this process, would the number of iterations be manageable before we hit a practical limit?

Kai: Well, they mention an outer layer with finitely many constraints and an inner layer identifying the most violated unitary constraint. They iterate until no constraint is violated or they hit a set maximum number of outer iterations Kmax.

Mira: That sounds like a practical way to handle the infinite constraints by using a finite approximation initially, and then refining it iteratively. It's about finding that projection point sigma star efficiently enough for testing.

Lev: But you have to be careful about those approximations; if the algorithm stops early, are we missing the true closest point or just a close enough one?

Paper summary: Kai: The paper says this systematic alternation empirically converges to the optimal solution sigma star, which then allows them to explicitly generate the optimal teleportation witness. They state that this construction ensures that Tr(W sigma) is non-negative for all states in S.

Mira: So the witness W they construct is guaranteed to have that specific property, which confirms its role as an optimal teleportation witness based on the projection geometry at sigma star.

Lev: And we know they also confirmed that Tr(W rho) is zero if and only if rho is useful for teleportation. That’s a strong necessary and sufficient condition, which is exactly what we want for experimental verification.

Kai: They test this method on three classes of states to show it works in practice. For Werner states, they reproduce results from reference sixteen, finding the critical visibility p equals one third for teleportation with their witness <ref:2605.22226#pg2>.

Mira: And for those two-qubit maximally entangled mixed states, their method yields the exact usefulness threshold, which they say existing inequality-based criteria in reference fifteen just can't get <ref:2605.22226#pg2,method yields the exact usefulness threshold, which>.

Lev: That’s significant because it means this geometric approach isn't just agreeing with old bounds; it’s finding a better one for these specific states. It suggests that existing criteria might be too loose or not tight enough in those regimes.

Kai: They also found a nonzero geometric distance at p equals zero point two five for the maximally entangled mixed states, which matches the threshold derived from the average teleportation fidelity criterion. That’s a nice cross-check between their method and another way of looking at it.

Mira: And for two-qutrit non-maximally entangled states mixed with a separable state, they found the transition point is pc equals zero point three seven five, and they extracted different optimal witnesses depending on the visibility p less than one. That shows flexibility in how the witness changes based on the input state.

Lev: So this paper gives us a concrete way to predict which states we can actually use for teleportation, and it seems robust enough across different entanglement structures tested so far. But what about states with more qudits?

Kai: The authors say this framework is designed to be extended to study multipartite entangled states once their geometric structure is fully characterized. They focus specifically on two-qudit states right now.

Mira: So the main implication here is that we have a general operational framework for teleportation usefulness using this shortest distance method, moving away from just relying on abstract inequalities.

Lev: It's a solid theoretical tool, but the real test will be running that two-layer cutting-plane algorithm on actual noisy, physical systems and seeing if those iteration counts stay low enough for a real experiment.

Kai: Yeah, that’s where the experimental verification comes in. The paper lays out the framework; now we have to see how much overhead it actually lands on the hardware we're building.

Conclusion: Kai: So we've seen how they build these optimal teleportation witnesses using geometry to solve a shortest distance problem to states that aren't useful for teleportation, and now we’re at the end of this paper called "Geometric Construction of Optimal Teleportation Witnesses" by Jia, Guo, Bai, Li, Dong.

Mira: Yeah so what does it actually mean when you put all that geometry together? It gives us a way to tell if a two-qudit entangled state is useful for teleportation using something operational instead of just some old inequalities.

Lev: Operational means we can actually try to build this thing on hardware, right? I'm thinking about how much overhead this whole geometric construction adds when you have real noise and error correction in the picture.

Kai: Exactly that's the thing with me, Lev, because if we want to run it on superconducting qubits, we need to know if the number of iterations they use for that two-layer cutting-plane algorithm is manageable before we start worrying about decoherence.

Mira: The paper claims this method gives us a necessary and sufficient criterion for usefulness by deriving the witness from the projection point sigma star, and that distance D of rho also acts as a test for whether it works.

Lev: If D of rho is greater than zero, that means the state is useful, but what does that actually tell us about the fidelity we can get on our actual quantum computer?

Kai: It tells us there's a non-zero gap in some sense, and the construction gives us a specific formula for the witness W based on that closest point sigma star.

Mira: And it sounds like this is a general framework, not just tied to one specific type of entanglement we were looking at before. It’s designed to handle arbitrary two-qudit entangled states.

Lev: That generality is cool, but I still wonder if the complexity of finding that closest point sigma star in the convex set S isn't going to make this too slow for real hardware implementation.

Kai: That's where we need to see if their iterative cutting-plane algorithm actually converges fast enough on a real chip, because if it takes too many steps, the geometric construction loses its practical value.

Mira: So the implication is that this geometric approach might be a way to get tighter bounds on which entangled states can actually perform quantum teleportation in practice.

Lev: Right, so it’s giving us a sharper tool to predict useful states, but we still gotta worry about how much computation it demands before we can actually build and measure these optimal witnesses.

Beijing University of Posts and Telecommunications School of Mathematical Sciences Key Laboratory of Mathematics and Information Networks State Key Laboratory of Networking and Switching Technology Beihang University School of Instrumentation Science and Opto-Electronics Engineering

quant-ph

Submitted: 2026-05-21

Updated: 2026-10-20

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 74/100

The gist: The gist The geometric construction of optimal teleportation witnesses provides an operational necessary and sufficient criterion for identifying the teleportation usefulness of arbitrary two-qudit

Key concepts

Useless States Set (S)
This is a convex set of quantum states that are useless for quantum teleportation. It is defined by constraints related to local unitary operations ($U$) and the trace of the product $\text{Tr}(\sigma U) \le 1/d$ for all $U$. States within this set cannot be used effectively in the teleportation process.
Projection Point ($\sigma^*)$
This is the unique state within the useless set S that is closest to a given target state $\rho$. Finding this point involves solving a shortest distance optimization problem. The paper uses this point to construct the optimal witness, which is central to determining usefulness.
Optimal Teleportation Witness (W)
This is a specific mathematical construct derived from the projection geometry at $\sigma^*$. It is defined as $W = n - cI$, where $n$ relates $\rho$ and $\sigma^*$. This witness serves as an operational tool; its properties (like $\text{Tr}(W\sigma) \ge 0$) provide the necessary and sufficient condition for a state's teleportation usefulness.

Summary

The gist The geometric construction of optimal teleportation witnesses provides an operational necessary and sufficient criterion for identifying the teleportation usefulness of arbitrary two-qudit entangled states by solving a shortest distance problem to the set of useless states.

Geometric Construction Method

The paper proposes a geometric method to construct optimal teleportation witnesses from the projection geometry at the unique closest state σ∗ ∈ S to a given state ρ, where S is the convex set of states useless for teleportation and D(ρ) itself serves as a necessary and sufficient condition for usefulness. The construction involves finding the projection point σ∗ ∈ S by solving the shortest distance optimization problem between ρ and S. This problem is formulated as min σ ρ − σF s.t. σ ≥ 0, Tr(σ) = 1, Tr(σPU) ≤ 1/d, ∀U ∈ U(d).

Two-Layer Cutting-Plane Algorithm

To solve the shortest distance problem D(ρ) = ρ − σ∗F explicitly, a two-layer iterative cutting-plane algorithm is developed. The outer layer solves a relaxed semidefinite program (SDP) to obtain a candidate state σ(k) by solving min σ ρ − σF s.t. σ ≥ 0, Tr(σ) = 1, Tr(σPU) ≤ 1/d, ∀U ∈ Uk. The inner layer searches for a unitary U that approximately maximizes Tr(σ(k)PU), and if the estimated value exceeds 1/d, the corresponding constraint is added to the outer problem. This procedure iterates until no constraint is violated or a maximum number of outer iterations Kmax is reached.

Criterion and Witness Construction

The constructed optimal teleportation witness W is derived from the projection point σ∗ and the optimal supporting hyperplane tangent to S at σ∗. Specifically, W = n − c I, where n = ρ − σ∗ and c = Tr(nσ∗)Tr(I). This construction ensures Tr(W σ) ≥ 0 for all σ ∈ S and Tr(W ρ) 0 is also confirmed as a necessary and sufficient condition for usefulness.

Application to Entangled States

The method was applied to three classes of entangled states to demonstrate its applicability. For two-qubit Werner states, the critical threshold is found at p = 1/3, and the optimal witness is consistently extracted. For two-qubit maximally entangled mixed states, the method detects a nonzero geometric distance at p = 0.25, which matches the threshold derived from the average teleportation fidelity criterion. In the case of two-qutrit non-maximally entangled states mixed with a separable state, the transition point is found at pc = 0.375, and the algorithm successfully extracts different optimal witnesses corresponding to different visibility values p < 1.

Conclusion

The geometric construction method provides a general and operational framework for studying the teleportation usefulness of arbitrary two-qudit entangled states by solving the shortest distance problem to the set S of useless states. This approach offers a necessary and sufficient criterion that surpasses existing inequality-based criteria in identifying useful states. The algorithm successfully tracks the sliding motion of the optimal supporting point σ∗ on the boundary of S, allowing it to construct corresponding optimal witnesses dynamically. This framework is designed to be extended to study multipartite entangled states once their geometric structure is characterized. The overall method provides a general and operational framework for studying the teleportation usefulness of arbitrary two-qudit entangled states. The geometric construction of witnesses has been used in entanglement theory.

--- Page 1 ---

Geometric Construction of Optimal Teleportation Witnesses

Yanning Jia,1, 2, 3 Fenzhuo Guo,1, 2, 3 Mengxuan Bai,1, 2, 3 Mengyan Li,1, 2, 3 Haifeng Dong.

--- Page 2 ---

not all entangled states are useful for quantum teleportation. We present a geometric method to construct optimal teleportation witnesses that provide operational necessary and sufficient criteria for identifying the teleportation usefulness of arbitrary two-qudit entangled states. Specifically, by developing a two-layer iterative cutting-plane algorithm to solve the shortest distance problem from the target state ρ to the convex set S of useless states, we obtain the projection point σ∗ ∈ S and then construct the optimal teleportation witness from the projection geometry. Moreover, the shortest distance D(ρ) obtained during this construction also serves as a necessary and sufficient criterion for usefulness.

--- Page 3 ---

The main challenge in the construction is finding this closest state σ∗, which we recast as the solution of a shortest distance optimization problem between ρ and the convex set S. This problem is nontrivial because S is defined by uncountably many local-unitary constraints, leading to a semi-infinite programming formulation.

--- Page 4 ---

If an entangled state ρ does not belong to S, then ρ is useful for teleportation. When ρ /∈ S, the projection theorem guarantees the existence of a unique closest point σ∗ ∈ S on the boundary of the convex set S, i.e., the projection of ρ onto S [22]. Define the normal vector n = ρ − σ∗, which points outward from S. Because σ∗ is the closest point, n satisfies the supporting hyperplane condition: Tr(ρ − σ∗)(σ − σ∗) ≤ 0, ∀σ ∈ S.

--- Page 5 ---

Thus W is a teleportation witness. Because it is derived from the projection point σ∗ and the optimal supporting hyperplane tangent to S at σ∗, it is called an optimal teleportation witness. Consequently, the key to constructing an optimal teleportation witness is to find the closest point σ∗ from ρ to the convex set S. To this end, we define the shortest distance D(ρ) = min σ ρ − σF, where AF = p Tr(A†A) is the Frobenius norm. Since S is compact, the minimum is attained, i.e., there exists σ∗ ∈ S such that D(ρ) = ρ − σ∗F [16] <ref:

Improvements for AI systems

  1. bold geometric criterion for teleportation usefulness extraction of optimal witness operator using shortest distance D(ρ). This allows AI systems to determine if an arbitrary entangled state ρ is useful for teleportation by checking if the "shortest distance D(ρ) > 0," providing a necessary and sufficient condition that surpasses existing inequality-based methods.

  2. two-layer cutting-plane algorithm for solving the shortest distance problem. The outer layer solves a relaxed problem containing only finitely many unitary constraints to obtain a candidate state σ(k), while the inner layer performs randomized fixed-point iteration with restarts to estimate the maximal entangled fraction F est max, enabling robust convergence to the optimal solution σ∗ and its corresponding optimal teleportation witness W.

  3. dynamic tracking of optimal witness orientation for complex states. For non-maximally entangled states, the method allows for the construction of different supporting points on the boundary as p varies, resulting in optimal witnesses that are also different, enabling AI to adapt and construct witnesses tailored to specific asymmetries in target states.

Related papers