From Permutation Symmetry to Communication Bounds and Additivity

arXiv:2610.02176 · quant-ph · Submitted 2026-10-01 · 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: I'm Kai, and with me are Mira and Lev, guest researcher.

Mira: Today's paper: "From Permutation Symmetry to Communication Bounds and Additivity".

Kai: Correlations across channel uses can improve quantum communication rates, making optimization over arbitrarily large blocks a central difficulty in determining quantum capacity.

Mira: First, who's behind it and why it matters.

Paper summary: Kai: So Mira, I'm really interested in this paper, "From Permutation Symmetry to Communication Bounds and Additivity." It claims that for any fixed finite-dimensional memoryless channel, optimizing over arbitrarily large blocks doesn't give you a consistent advantage when you have full permutation invariance.

Mira: That sounds like it tackles the central difficulty in figuring out quantum capacity—how correlations across uses can improve things, but this paper shows how full permutation invariance restricts that advantage. It basically argues that for every fixed channel, the optimized coherent information per use ends up converging to a single-use maximum because of this symmetry.

Lev: From my side, I'm thinking about what this means for actual hardware implementation; if the asymptotic optimization just hits the single-use maximum, then running very long blocks might not yield significantly better performance than a single use anyway, which simplifies things for error correction codes we have to actually build.

Kai: Exactly. And the paper shows that even though you can't get that asymptotic advantage, there's still a possibility of finite-block superadditivity, which is an important distinction we need to track when designing experiments.

Mira: Right, and they quantify this limitation by defining q sym,n(N):= one/n rho A n I c(N n, rho A n), which is then bounded between Q(one)(N) and Q(one)(N) + O(n/sqrt n) for any fixed channel <ref:2610.02176#pg0>.

Lev: That bound suggests that while you can't get the asymptotic benefit of using a long block, you still have some control over how much better you can do with a finite block length, which is something we need to model carefully when designing codes that rely on correlation across uses.

Kai: And the paper also mentions a strong converse for pure-source entanglement generation codes with permutation-invariant inputs, stating that at any fixed rate above Q(one)(N), the fidelity tends to zero as the block length grows <ref:2610.02176#pg0,a strong converse for pure-source entanglement generation>.

Mira: That's a significant constraint, showing that when you restrict your input structure by symmetry, you can establish upper bounds on performance for entanglement generation codes even when they are doing better than the single-use rate.

Lev: If we were trying to run this on real hardware with error correction, this strong converse bound would tell us exactly where we need to set our expectations regarding fidelity as we scale up the block size.

Paper summary: Kai: Moving on, let's look at the structural framework they use to connect these ideas—they identify a common structure linking completely bounded norms, channel discrimination against a replacer, and entanglement-assisted communication through those "sandwiched Rényi channel quantities."

Mira: That framework seems central because it allows them to derive systematic proofs for multiplicativity and additivity results across these seemingly different areas by employing De Finetti reduction and permutation covariance.

Lev: I wonder how this variational setup translates into practical tools for analyzing error correction codes; does this common framework offer a unified way to analyze the performance of different types of correlated inputs?

Kai: The paper then applies postselection to these common functionals to establish additivity results, showing that for a fixed output weight, De alpha(N one N two R sigma one R sigma two) = De alpha(N one R sigma one) + De alpha(N two R sigma two) <ref:2610.02176#pg0>.

Mira: That additivity for the channel divergence, and similarly for the channel Rényi mutual information, I e alpha(N one N two) = I e alpha(N one) + I e alpha(N two), is quite powerful because it shows that these quantities behave nicely when you consider tensor products of inputs <ref:2610.02176#pg0>.

Lev: For error correction, having the mutual information add linearly across uses or input blocks would be fantastic for analyzing how redundancy scales, and this result suggests a level of structural consistency we haven't seen before in this context.

Kai: The authors also tie minimum output entropy to this common functional by defining h sym a,n(N):= tau n in D(A n) P A(pi) tau n P A(pi) = tau n and proving its limit as one/n h sym a, n(N) = h alpha(N) <ref:2610.02176#pg0>.

Mira: That convergence to the single-use minimum, h alpha(N), is a key finding because it complements the failure of unrestricted minimum output entropy additivity. It shows that even under symmetry constraints, you can't sustain an asymptotic reduction in output entropy per use below that single-use minimum.

Lev: So, if we think about the physical manifestation of this, it implies that correlations across many uses don't fundamentally help reduce the inherent uncertainty per use below what a single use can achieve when symmetry is involved.

Kai: And the final discussion reinforces this by confirming that full permutation invariance constrains coherent-information advantage, meaning asymptotic optimization simplifies down to just one channel use.

Mira: Plus, they solidify the idea that for pure-source entanglement generation codes with permutation-invariant inputs, a strong converse bound is established at rates above Q(one)(N), which sets clear limits on how well these codes can perform <ref:2610.02176#pg0,for pure-source entanglement generation>.

Paper summary: Lev: It sounds like the practical implication here is that when we design protocols relying on highly correlated inputs, we need to account for this symmetry restriction early on because it dictates the asymptotic behavior of the coherent information gains.

Kai: So, to sum up what we've covered in "From Permutation Symmetry to Communication Bounds and Additivity," the main thrust is that full permutation invariance limits coherent information optimization so that it converges to a single-use maximum, even while finite-block superadditivity remains possible.

Mira: And the paper provides a very robust structural framework using common variational formulas and De Finetti reduction to prove additivity for several important quantities, linking completely bounded norms and channel divergence estimates in a systematic way.

Lev: For the real world of quantum communication, this means we have a much better handle on where the limits are when we introduce these kinds of input symmetries into our designs for error correction and capacity analysis.

Kai: It really paints a clearer picture of how symmetry acts as a constraint on what correlations can actually achieve in terms of asymptotic communication rates.

Mira: And it extends beyond just coherent information, showing that this symmetry structure applies to other quantities like channel discrimination against a replacer, which is important for understanding the broader landscape of quantum information theory.

Lev: I think the most tangible impact is providing those rigorous bounds for entanglement generation codes, which directly informs how we set fidelity targets when we try to generate entangled states using these symmetric input structures.

Kai: So, to wrap up on this paper, "From Permutation Symmetry to Communication Bounds and Additivity," it shows that symmetry restricts the asymptotic advantage in coherent information optimization while providing a consistent mathematical language—through sandwiched Rényi quantities and De Finetti reduction—to prove additivity for related channel measures.

Mira: And these results give us precise bounds on what we can expect from communication rates and entanglement fidelity when inputs are restricted by permutation invariance, confirming that asymptotic optimization reduces to a single channel use under those conditions.

Lev: It’s a solid piece of theoretical work because it connects the abstract symmetry to concrete performance limits, giving us tools to assess the feasibility of using highly correlated inputs in large-scale quantum systems.

Conclusion: Kai: So, we've looked at the heavy math behind "From Permutation Symmetry to Communication Bounds and Additivity," where they really dig into how input structure affects quantum communication rates, and now we need to talk about what this actually means for us in simple terms.

Mira: I think the core idea is that when you impose full permutation symmetry on your inputs, it puts a specific ceiling on how much better you can get from using many uses of a channel compared to just one use.

Lev: From an error correction standpoint, if this holds true, it tells us that simply stacking more uses of a correlated input might not give us the asymptotic improvement we're hoping for in our large-scale codes.

Kai: Exactly, and the authors are proving this limitation by showing that their optimized performance converges to a single use maximum with some specific logarithmic correction term.

Mira: That convergence result is interesting because it shows that even though we can't get an advantage from long blocks in the limit, there's still a controlled way we can quantify how close we are to that single-use rate.

Lev: I mean, for hardware, this suggests that if we design protocols based on these highly symmetric inputs, we might need to be careful about expecting superadditivity when scaling up the block length.

Kai: So what does this mean for the broader field? The paper establishes a very specific mathematical boundary imposed by symmetry on coherent information and entanglement generation.

Mira: It provides a unified variational framework that connects concepts like completely bounded norms with things like channel divergence, which is really neat because it links different theoretical areas together.

Lev: That shared framework might be useful later when we try to analyze the performance of real quantum error-correcting codes under these types of symmetry constraints.

Kai: Indeed, and this whole paper sets a clear benchmark for what's achievable under these specific input conditions for capacity and entanglement fidelity.

Mira: It really shows how deeply rooted symmetry can be in constraining the fundamental limits of quantum communication protocols.

Lev: So the big picture here is that we have a sharper tool to analyze performance bounds when dealing with structured inputs, which is something we need for practical implementation.

Kai: And once we understand these precise constraints, it opens up new avenues for designing codes that respect those symmetries while still performing well.

Zahra Baghali Khanian, Debbie Leung, Graeme Smith

Perimeter Institute for Theoretical Physics · Institute for Quantum Computing, University of Waterloo

quant-ph

Submitted: 2026-10-01

Updated: 2026-10-01

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

Importance score: 91/100

The gist: Correlations across channel uses can improve quantum communication rates, making optimization over arbitrarily large blocks a central difficulty in determining quantum capacity.

Key concepts

qsym_n(N)
This quantity measures the optimized coherent information per use under permutation invariance. The authors show that this value is tightly bounded by the single-use maximum, indicating that asymptotic optimization requires only one channel use.
Deα(N∥Rσ)
This represents a common functional used to express channel divergence against a replacer. It forms the basis of a shared variational framework that allows the authors to systematically prove additivity results for various quantum quantities.
h_sym^a,n(N)
This term quantifies the minimum output entropy under symmetry constraints. It converges to a single-use minimum as block length increases, showing that optimized entropy per use is limited by the single-use case.

Terminology

Summary

Correlations across channel uses can improve quantum communication rates, making optimization over arbitrarily large blocks a central difficulty in determining quantum capacity. This work investigates how full permutation invariance limits this advantage for every finite-dimensional memoryless channel, showing that asymptotic optimization reduces to a single channel use while retaining the possibility of finite-block superadditivity.

The gist

For every fixed finite-dimensional channel, the optimized coherent information per use converges to the single-use maximum and consequently attains its supremum at a finite block length.

Limitations on Coherent Information Bounds

The paper quantifies the limitation imposed by permutation invariance on coherent information optimization. The authors define a quantity, denoted as:

qsym n(N):= 1/n max ρA n Ic(N⊗n, ρA n)

They prove that for every fixed finite-dimensional channel, the following holds:

  1. Q(1)(N) ≤ qsym n(N) ≤ Q(1)(N) + O(log n/√n).

  2. The lower bound follows from product inputs, whereas the upper bound is uniform over the entire symmetric input class.

This demonstrates that the asymptotic coherent-information optimization under full permutation invariance requires only a single channel use, while finite-block superadditivity remains possible but its excess per use vanishes at a controlled rate. Furthermore, for pure-source entanglement-generation codes with fully permutation-invariant channel inputs, they establish a strong converse bound: at any fixed communication rate above Q(1)(N), the fidelity tends to zero as the block length grows.

Structural Framework for Additivity and Multiplicativity

The authors identify a common structure underlying completely bounded norms, channel discrimination against a replacer, and entanglement-assisted communication through sandwiched Rényi channel quantities. They derive systematic alternative proofs of known multiplicativity and additivity results within a shared variational framework by employing De Finetti reduction and permutation covariance. This framework involves:

  1. Minimum output Rényi entropy fitting into the framework through the pure-input restriction of the variational formula for the completely bounded norm, followed by a logarithmic rescaling.

  2. The common Choi representation, where channel divergence is expressed as: Deα(N∥Rσ) = α/(α − 1) log max ρ∈D(A′) ωα(ρ, σ, N).

Additivity Results for Channel Quantities

The paper applies the postselection method to these common functionals to establish additivity results. For a fixed output weight, they show:

Deα(N1 ⊗ N2∥Rσ1 ⊗ Rσ2) = Deα(N1∥Rσ1) + Deα(N2∥Rσ2)

This equality is derived by applying Theorem 10 to the maps and multiplying its logarithm by α/(α − 1). Similarly, for the channel Rényi mutual information, they prove:

Ieα(N1 ⊗ N2) = Ieα(N1) + Ieα(N2)

Control over Minimum Output Entropy

The authors relate minimum output entropy to the common functional. They define:

h sym a,n(N):= min τ n∈D(A n) P A(π) τ n P A(π)† = τ n for all π∈Sn.

They prove the convergence of this quantity to the single-use minimum:

lim n→∞ 1/n h sym a,n (N) = hα(N)

This symmetry-restricted conclusion complements the failure of unrestricted minimum output entropy additivity, showing that the optimized entropy per use converges to the single-use minimum.

Implications for Capacity Bounds

The results show that for permutation-invariant inputs, correlations cannot sustain an asymptotic reduction in output entropy per use below the single-use minimum. This entropy limit holds for Rényi orders greater than one and for the von Neumann entropy, extending the role of symmetry from communication bounds to the control of output entropy. The final discussion confirms that full permutation invariance constrains coherent-information advantage, with asymptotic optimization reducing to a single channel use. For pure-source entanglement generation codes with permutation-invariant channel inputs, a strong converse bound is established at rates above Q(1)(N).

Summary of Key Findings

The paper establishes that for permutation-invariant inputs, the optimized coherent information per use converges to Q(1)(N) with a correction of order O(log n/√n). It also provides asymptotic bounds for entanglement generation codes under this symmetry constraint. Crucially, it identifies a common variational framework using the functional ωα to prove multiplicativity and additivity for completely bounded norms, channel divergence against a replacer, and channel Rényi mutual information.

Improvements for AI systems

Based on the provided scientific paper, here are specific improvements that could be made to AI systems by leveraging these theoretical results:

  1. Improve Quantum Communication/Networking Protocols: The paper establishes bounds on coherent information for permutation-invariant inputs and provides strong converse bounds for entanglement generation codes. An AI system designed for quantum networking or secure communication protocols could use this to design codes and input strategies that maximize the achievable quantum rate over a finite block length, ensuring the system operates near the asymptotic single-use capacity limit.

  2. Optimize Quantum Error Correction Codes: The paper provides bounds on minimum output Rényi entropy for permutation-invariant inputs and strong converse bounds for entanglement generation codes. An AI system could be used to design quantum error correction codes (like those based on repetition or graph codes) that are specifically optimized to maintain a high fidelity against noise, even when the input states exhibit permutation symmetry.

  3. Enhance Quantum Channel Modeling and Discrimination: The paper identifies common structural features underlying completely bounded norms, channel discrimination against a replacer, and entanglement-assisted communication via sandwiched Rényi channel quantities. An AI system could use these common functional frameworks to develop more robust models of quantum channels, allowing for better discrimination between a true channel and an eavesdropping replacer in noisy environments.

  4. Develop Quantum Information Processing for Correlated Inputs: The results on additivity and multiplicativity of the completely bounded norm, channel Rényi mutual information, and entanglement-assisted communication allow an AI system to systematically analyze how correlations across multiple uses improve performance. This enables the design of complex quantum processing tasks that leverage structured, symmetric input correlations to achieve better performance than independent inputs, specifically targeting improved classical communication rates or entanglement distillation/generation.

  5. Improve Quantum Capacity Estimation: The paper quantifies the correction term for permutation-invariant inputs to coherent information as being of order O(log n/√n). An AI system could be used in capacity estimation algorithms to precisely predict and account for this finite-block length correction, leading to more accurate estimations of the true quantum capacity when using structured or symmetric input sources.

  6. Optimize Minimum Output Entropy for Quantum Data Compression/Coding: The results on minimum output Rényi entropy (including the convergence to single-use minimum) allow an AI system to design quantum compression schemes or coding techniques that minimize the information loss (output entropy) when dealing with permutation-invariant quantum data, ensuring that the achieved compression rate approaches the fundamental single-use limit as more data is processed.

Sources

Related papers