When can local unitaries generate a complete orthogonal basis?

arXiv:2611.16921 · quant-ph, math-ph, math.MP · Submitted 2026-11-12 · 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.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "When can local unitaries generate a complete orthogonal basis?".

Mira: Here's the gist of the paper. When can local unitaries generate a complete orthogonal basis?

Kai: The question of local encodability asks: can a given, multipartite pure quantum state be completed to an orthonormal basis using only local unitary rotations?

Mira: The paper addresses the problem of universal local encodability (ULE), which asks whether a given multipartite pure quantum state can be completed to an orthonormal basis using only local unitary rotations, a feature that has applications in quantum information theory, engineering and math The question of local encodability is formally posed as whether the two-sided unitary orbit of an arbitrary matrix contains a Hilbert-Schmidt orthonormal matrix basis For the case of bipartite systems, local encodability is equivalent to asking whether the two-sided unitary orbit of an arbitrary matrix contains a Hilbert-Schmidt orthonormal matrix basis The state can be associated with a complete projective measurement in which every basis state has the same local-unitary invariants and therefore the same entanglement as the original state Heuristics and Dimensional Constraints A heuristic argument suggests that universal local encodability should be possible if and only if a certain inequality involving degrees of freedom is fulfilled, specifically Xn j=1 d 2j!

Kai: ≥ Yn j=1 dj + n This heuristic predicts ULE in any bipartite system, and the paper shows that it indeed holds for many choices of local dimensions However, this heuristic predicts that in many multipartite systems, ULE should not hold The paper also provides a necessary condition for universality based on a heuristic parameter count which excludes universal local encodability in infinitely many cases Constructive Methods and Bipartite Results The authors resolve the open conjecture by proving universality for two qutrits, and they construct universal local encodings in dimensions d×2 for d ≥ 2, in dimensions d×3 for d = 3, 4 and d ≥ 6, and in every bipartite system with one or two infinite-dimensional subsystems They establish ULE in ∞ × dB systems for all dB using the lifting lemma and Hurwitz-Radon matrices Furthermore, they show that for each fixed dB, there exists a finite dA ≥ dB such that dA ×dB is universally locally encodable using a Hurwitz-Radon construction Operational Interpretation and No-Go Theorems The paper gives universal local encodability an operational interpretation by showing that non-ULE is equivalent to whether general measurements yield an advantage over projectively simulable ones in discrimination tasks where unknown, random local unitary rotations have to be detected The authors provide explicit constructions showing ULE for infinitely many bipartite system dimensions, including dA × 2 for all dA ≥ 2, dA × 3 for all dA ≥ 3 except of dA = 5 and ∞ × dB for all dB They generalize non-existence arguments to many multipartite systems provided their local dimensions violate a naive degree-of-freedom counting inequality The final result is the impossibility theorem, Theorem V.4, which states that if Xn j=1 d 2j < (n − 1) + Yn j=1 dj holds, then H is not universally locally encodable The paper concludes by showing that universal local encodability is possible for two qutrits, Theorem C.2 It also demonstrates that every state is locally encodable for two parties and local dimension dA ≥ 2 and dB = 2, Proposition B.10 The operational interpretation links ULE to the existence of an optimal projective measurement for a class of invariant estimation tasks Future research directions include generalizing findings to multipartite systems with at least one infinite dimensional part The authors also explore the structure of groups that generate maximally coherent locally encodable states, noting that for coprime dimensions in bipartite systems, the only groups satisfying this structure are cyclic Finally, they provide an explicit construction for 3 × 3 systems using a group approach The paper is organized as follows.

Paper summary: Kai: So, this paper lays out that there's a certain inequality involving degrees of freedom that seems to predict whether ULE should be possible in many cases >

Mira: The main point is that if that inequality isn't met for your local dimensions, then the state simply cannot be universally locally encodable, and they show this impossibility theorem based on Theorem V.four > <ref:2611.16921#pg1>

Lev: It's interesting because this links a simple counting argument to deep properties of quantum states; if we can prove it fails for a certain setup, that tells us something concrete about the limits of what local operations can achieve on that hardware >

Conclusion: Kai: So, looking at the whole paper, the main thing is that they resolved an open conjecture by proving universality for two qutrits >

Mira: They also showed that even for bipartite systems, there are constructions available for many dimensions and that every state is locally encodable if you fix a local dimension of two or three parties >

Lev: What this means practically is that it helps us figure out the limits of quantum information tasks; knowing when something *can't* be done with just local operations tells us exactly where we need to focus our error correction efforts >

Institut für Theoretische Physik III, Heinrich-Heine-University Düsseldorf

quant-ph, math-ph, math.MP

Submitted: 2026-11-12

Updated: 2026-11-12

Comments: 9 + 18 pages, 2 figures

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

Importance score: 73/100

The gist: When can local unitaries generate a complete orthogonal basis? The question of local encodability asks: can a given…

Key concepts

Universal Local Encodability (ULE)
This asks if any given multipartite pure quantum state can be transformed into a complete orthonormal basis using only local unitary operations applied independently to each subsystem. It relates the state's structure to the set of matrices reachable through local rotations.
Local Unitary Orbit
This refers to the set of all possible quantum states that can be obtained from an initial state by applying a combination of unitary transformations, where each transformation acts only on one specific subsystem. ULE is determined by whether this orbit contains a complete orthonormal basis.
Hurwitz-Radon Construction
This is a specific mathematical method used to prove universal local encodability in certain dimensions. It provides explicit constructions showing that for specific system sizes, there exists a finite set of local unitary operations that can generate the entire required basis.

Terminology

Summary

When can local unitaries generate a complete orthogonal basis? p. 0 The question of local encodability asks: can a given, multipartite pure quantum state be completed to an orthonormal basis using only local unitary rotations? p. 0

Introduction and Problem Formulation

The paper addresses the problem of universal local encodability (ULE), which asks whether a given multipartite pure quantum state can be completed to an orthonormal basis using only local unitary rotations, a feature that has applications in quantum information theory, engineering and math The question of local encodability is formally posed as whether the two-sided unitary orbit of an arbitrary matrix contains a Hilbert-Schmidt orthonormal matrix basis For the case of bipartite systems, local encodability is equivalent to asking whether the two-sided unitary orbit of an arbitrary matrix contains a Hilbert-Schmidt orthonormal matrix basis The state can be associated with a complete projective measurement in which every basis state has the same local-unitary invariants and therefore the same entanglement as the original state

Heuristics and Dimensional Constraints

A heuristic argument suggests that universal local encodability should be possible if and only if a certain inequality involving degrees of freedom is fulfilled, specifically Xn j=1 d 2j! ≥ Yn j=1 dj + n This heuristic predicts ULE in any bipartite system, and the paper shows that it indeed holds for many choices of local dimensions However, this heuristic predicts that in many multipartite systems, ULE should not hold The paper also provides a necessary condition for universality based on a heuristic parameter count which excludes universal local encodability in infinitely many cases

Constructive Methods and Bipartite Results

The authors resolve the open conjecture by proving universality for two qutrits, and they construct universal local encodings in dimensions d×2 for d ≥ 2, in dimensions d×3 for d = 3, 4 and d ≥ 6, and in every bipartite system with one or two infinite-dimensional subsystems They establish ULE in ∞ × dB systems for all dB using the lifting lemma and Hurwitz-Radon matrices Furthermore, they show that for each fixed dB, there exists a finite dA ≥ dB such that dA ×dB is universally locally encodable using a Hurwitz-Radon construction

Operational Interpretation and No-Go Theorems

The paper gives universal local encodability an operational interpretation by showing that non-ULE is equivalent to whether general measurements yield an advantage over projectively simulable ones in discrimination tasks where unknown, random local unitary rotations have to be detected The authors provide explicit constructions showing ULE for infinitely many bipartite system dimensions, including dA × 2 for all dA ≥ 2, dA × 3 for all dA ≥ 3 except of dA = 5 and ∞ × dB for all dB They generalize non-existence arguments to many multipartite systems provided their local dimensions violate a naive degree-of-freedom counting inequality The final result is the impossibility theorem, Theorem V.4, which states that if Xn j=1 d 2j < (n − 1) + Yn j=1 dj holds, then H is not universally locally encodable

Conclusion and Future Directions

The paper concludes by showing that universal local encodability is possible for two qutrits, Theorem C.2 It also demonstrates that every state is locally encodable for two parties and local dimension dA ≥ 2 and dB = 2, Proposition B. p. 110 The operational interpretation links ULE to the existence of an optimal projective measurement for a class of invariant estimation tasks Future research directions include generalizing findings to multipartite systems with at least one infinite dimensional part The authors also explore the structure of groups that generate maximally coherent locally encodable states, noting that for coprime dimensions in bipartite systems, the only groups satisfying this structure are cyclic Finally, they provide an explicit construction for 3 × 3 systems using a group approach The paper is organized as follows. p. 1 p. 7

Appendix Details

The appendix provides technical details for the Hurwitz-Radon Construction, showing that universal local encodability holds true for two harmonic oscillators (i.e., dA = dB = ∞) using Theorem III. p. 72 It also details topological constructions, showing that the set Td is compact, contractible and its boundary is given by the trace vectors The explicit construction for 3 × 3 systems involves complex matrix forms (C7). p. 4 The proof of the no-go theorem generalizes the criterion to all local dimension tuples where non-ULE was already suggested based on the violation of a heuristic dimensional inequality The proof of Lemma VI. p. 72 shows that for a state ψ⟩ not locally encodable, there exists an ε > 0 such that Rε POVM(ψ⟩⟨ψ) < Rε PVM(ψ⟩⟨ψ). p. 7 The result is summarized in Tables I and II. p. 1 The authors thank Raphael Brinster, Dagmar Bruß, Otfried G¨uhne and Tulja Varun Kondra for fruitful discussions. p. 7 Showing equivalence of universal local encodability with optimality of projective measurements in discrimination tasks connects regularity of a state to its stabilizer and some parts of the constructions. p. 7 The paper is organized as follows.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Universal Local Encodability Check: The system can determine if a given multipartite pure quantum state is universally locally encodable by checking if there exists an optimal projective measurement in a class of locally invariant estimation tasks. This involves verifying that for all states and all epsilon, Repsilon POVM(rho0) equals Repsilon PVM(rho0).

  2. Optimal Measurement Identification: The system can identify the minimum average loss over all possible POVM measurements for any probe state and invariant loss function by finding the optimal projective measurement, which is guaranteed if universal local encodability holds. This involves computing Dcmin, where Dcmin is the smallest eigenvalue of C and |eta> is a corresponding normalized eigenvector.

  3. State Discrimination Advantage Detection: The system can detect if non-ULE (non-universally locally encodable) states yield an advantage over projectively simulable ones in discrimination tasks by checking if general measurements yield an advantage over projectively simulable ones in discrimination tasks where unknown, random local unitary rotations have to be detected.

Sources

Related papers