Strict Hierarchy for Quantum Channel Certification to Unitary

Abstract: We consider the problem of quantum channel certification to unitary, where one is given access to an unknown $d$-dimensional channel $\mathcal{E}$, and wants to test whether $\mathcal{E}$ is equal to a target unitary channel or is $\varepsilon$-far from it in the diamond norm. We present optimal quantum algorithms for this problem, settling the query complexities in three access models with increasing power. Specifically, we show that: (i) $Θ(d/\varepsilon^2)$ queries suffice for incoherent access model, matching the lower bound due to Fawzi, Flammarion, Garivier, and Oufkir (COLT 2023). (ii) $Θ(d/\varepsilon)$ queries suffice for coherent access model, matching the lower bound due to Regev and Schiff (ICALP 2008). (iii) $Θ(\sqrt{d}/\varepsilon)$ queries suffice for source-code access model, matching the lower bound due to Jeon and Oh (npj Quantum Inf. 2026). This demonstrates a strict hierarchy of complexities for quantum channel certification to unitary across various access models.

• The paper establishes a strict complexity hierarchy with incoherent, coherent, and source-code models yielding Θ(d/ε²), Θ(d/ε), and Θ(√d/ε) query complexities, respectively.
• It introduces optimal certification algorithms leveraging techniques like the Choi–Jamiołkowski isomorphism and quantum amplitude estimation.
• Tight lower bounds and detailed implications refine our understanding of quantum certification versus full channel tomography.

Strict Complexity Hierarchy for Quantum Channel Certification to Unitary

Problem Overview

The paper "Strict Hierarchy for Quantum Channel Certification to Unitary" (2604.26900) addresses the quantum property testing problem of certifying whether a channel $\mathcal{E}$, accessed as a black box, is a specified unitary channel $\mathcal{U}$ or is at least $\varepsilon$-distant from $\mathcal{U}$ in the diamond norm. The core contribution is a matching upper and lower bound on the sample complexity in three access models—incoherent, coherent, and source-code—thereby establishing a strict hierarchy of quantum query complexity for this property testing task.

Access Models and Complexity Hierarchy

The complexity of quantum channel certification crucially depends on the power endowed by the access model:

• Incoherent Access: The algorithm is forced to measure after each call to $\mathcal{E}$ with no retention of quantum memory across queries. Optimal sample complexity is $\Theta(d/\varepsilon^2)$.
• Coherent Access: Full quantum memory and coherence are preserved, allowing entangled queries and joint measurements. The sample complexity reduces to $\Theta(d/\varepsilon)$.
• Source-Code Access: The algorithm receives a unitary implementation ("source code") $W$ for $\mathcal{E}$, with ability to invoke $W$ and $\mathcal{U}$0. The complexity drops further to $\mathcal{U}$1.

This demonstrates strict asymptotic separations across access models.

Technical Contributions

Optimal Algorithms in All Models

For each access model, the paper provides an explicit certification algorithm:

Incoherent Access:

Uses the Choi–Jamiołkowski isomorphism to relate the problem to discriminating entanglement fidelities. The critical technical link is a quantitative relationship between entanglement fidelity $\mathcal{U}$2 and the diamond norm $\mathcal{U}$3:

$\mathcal{U}$4

Estimating the entanglement fidelity with sufficient accuracy requires $\mathcal{U}$5 queries, and this bound is proven tight.

Coherent Access:

Amplifies the gap through sequential composition. By noting that iterating the map $\mathcal{U}$6 linearly increases the diamond-norm distance in the relevant regime, the difference is boosted and then efficiently detected. The algorithm bootstraps the incoherent test, allowing the certification with $\mathcal{U}$7 queries.

Source-Code Access:

Maps the problem into quantum amplitude estimation. The existence of the source code allows the construction of a controlled circuit whose amplitude encodes the entanglement fidelity gap. Amplitude estimation reduces the sample complexity to $\mathcal{U}$8, matching known lower bounds from Grover-like collision lower bounds generalized to quantum channels.

Tight Lower Bounds

For each model, matching lower bounds are presented or referenced:

• The incoherent bound follows from [FFGO23].
• The coherent bound is derived from a reduction to the $\mathcal{U}$9-faulty Grover problem [RS08], leveraging adversarial channels.
• The source-code lower bound is achieved by relating the problem to unitary identity testing [JO26].

Implications

This work provides several important theoretical advances:

1. Strict Model Separation: It is proven that additional quantum resources—coherence, and finally, access to the unitary realization—provably and strictly reduce query complexity for channel certification.
2. Tight Quantification: Not only are asymptotic exponents strictly separated, but the paper also determines the optimal polylogarithmic factors.
3. Algorithmic Template: The methods combine property testing with quantum information tools (Choi isomorphism, amplitude estimation, and diamond-norm continuity bounds), yielding transferable techniques for quantum property testing.
4. Optimality for Certification vs. Tomography: The results refine the understanding of the sample complexity of certification (decision problems) versus full channel tomography, which is far more resource-intensive.

Potential Future Directions

• Generalization to Other Properties: Extending this strict hierarchy to more general quantum learning and testing problems, including certifying other channel classes or continuous property predicates.
• Hierarchy Collapse Cases: Identifying problems where the hierarchy collapses (i.e., access to source code or coherence does not provide additional power).
• Finer Granularity in Access Models: Analyzing more sophisticated or intermediate access models, potentially enriching the hierarchy or resulting in more nuanced tradeoffs.
• Practical Realization: Investigating implementations under realistic noise and error models, considering the practical overheads of coherent or source-code access.

Conclusion

The results in "Strict Hierarchy for Quantum Channel Certification to Unitary" (2604.26900) rigorously establish that quantum channel certification to a unitary target yields a strict complexity hierarchy: $\varepsilon$0 (incoherent), $\varepsilon$1 (coherent), and $\varepsilon$2 (source code). The algorithms are optimal, the bounds are tight, and the work forms a benchmark for future studies in quantum property testing and learning theory in the presence of varying quantum query resources.