⚛️ quantum physics

Non-Gaussianity from superselection rules

This paper reinterprets non-Gaussianity and stellar rank as witnesses of particle entanglement under superselection rules and generalizes the stellar rank to arbitrary computational bases, thereby establishing it as a basis-dependent measure of genuine bosonic resources for quantum advantage.

Original authors: Nicolas Moulonguet, Eloi Descamps, José Lorgeré, Astghik Saharyan, Arne Keller, Pérola Milman

Published 2026-03-24

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Nicolas Moulonguet, Eloi Descamps, José Lorgeré, Astghik Saharyan, Arne Keller, Pérola Milman

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Why "Weird" Light Matters

Imagine you are trying to build a super-fast computer using light (photons). To make this computer beat our best classical supercomputers, the light needs to be "weird." In physics, we call this non-Gaussianity.

Think of "Gaussian" light as a calm, predictable lake. It's smooth, round, and follows standard rules. You can simulate this kind of light easily on a regular laptop. But to get a quantum advantage (the "superpower"), you need non-Gaussian light. This is like a lake with violent, jagged waves, or a stormy sea. It's chaotic, unpredictable, and impossible to simulate easily.

For a long time, scientists thought this "weirdness" (non-Gaussianity) was just a fancy way of saying "we added extra photons (particles of light) to the mix." They thought it was a simple recipe: Start with smooth light, add a few particles, and boom—you have a quantum resource.

This paper says: "Not so fast."

The authors argue that this "weirdness" isn't just about adding particles. It's actually a sign of entanglement (a deep, spooky connection) between the particles themselves, which only becomes visible when you treat the "clock" or "reference frame" of the light as a quantum object, not just a background setting.

The Core Analogy: The Orchestra and the Conductor

To understand the paper's breakthrough, let's use an analogy of an Orchestra.

1. The Old View (The "Single-Mode" View)

Imagine you are listening to a solo violinist. In the old way of thinking, we treat the violinist as the only thing that matters. We ignore the rest of the orchestra and the conductor.

The Problem: If the violinist plays a complex, "weird" note (non-Gaussian), we assume it's just because they are playing a difficult solo. We don't ask why they are playing it that way.
The Paper's Insight: The authors say, "Wait a minute! The violinist isn't playing in a vacuum. They are playing in an orchestra with a conductor."

2. The New View (The "Superselection" View)

In this paper, the "conductor" is the phase reference. In quantum physics, you can't measure the "phase" (the timing or rhythm) of light without a reference clock. Usually, scientists pretend this clock is perfect and classical (like a metronome that never ticks wrong).

The authors say: "No, the clock is also quantum!"

When you treat the clock (the reference) as a real quantum particle that can be entangled with the violinist (the light), a new picture emerges:

The "weird" notes the violinist plays are actually a result of the violinist and the conductor being entangled.
If they are perfectly synchronized (separable), the music is smooth and Gaussian (boring for quantum computing).
If they are entangled, the music becomes "weird" (non-Gaussian).

The Takeaway: The "weirdness" of the light isn't just a property of the light itself; it's a witness that the light is entangled with its own reference frame.

The "Stellar Rank": Counting the Stars

The paper focuses on a specific tool called the Stellar Rank.

The Metaphor: Imagine the state of the light is a constellation in the sky. The "Stellar Rank" is simply the number of stars in that constellation.
The Old Idea: Scientists thought you could have an infinite number of stars, and more stars meant more "quantum power."
The Paper's Discovery: The authors looked at how these constellations form when you zoom out from a tiny, finite system (a few particles) to a huge, continuous system (infinite light).

They found a surprising limit:

Even though the sky looks infinite, the "Stellar Rank" (the number of meaningful stars) is actually tiny compared to the total number of particles.
It's like looking at a galaxy from Earth. You see millions of stars, but the "Stellar Rank" is only counting the specific, bright stars that define the shape of the constellation. The rest are just background noise.

Why does this matter?
It means that for a state to be useful for quantum computing, it doesn't need to be "maximally weird." It just needs a specific, small number of "stars" (roots of a polynomial) that indicate it is entangled with its reference.

The "Basis" Problem: It Depends on How You Look

Here is the final twist. The paper argues that whether light looks "weird" (non-Gaussian) or "normal" (Gaussian) depends entirely on how you choose to measure it.

The Analogy: Imagine a sculpture.
- If you look at it from the front, it looks like a perfect circle (Gaussian).
- If you look at it from the side, it looks like a jagged square (Non-Gaussian).
- The object hasn't changed; your perspective (the "computational basis") has.

The authors show that the "Stellar Rank" is not an absolute number. It changes depending on which "lens" (basis) you use to view the quantum state.

If you use the standard "quadrature" lens (the usual way we look at light), you get one Stellar Rank.
If you change the lens to a different "computational basis," a state that looked normal might suddenly look "weird" and gain a high Stellar Rank.

The Conclusion: To truly know if a quantum state is powerful enough to beat a classical computer, you can't just look at it one way. You have to check if it has "weirdness" (entanglement) relative to the specific way you plan to use it.

Summary: What Did They Actually Prove?

The Origin of "Weirdness": Non-Gaussianity (the "weird" light needed for quantum computers) is actually a sign of particle entanglement between the light and its reference frame. It's not just "adding particles"; it's about how those particles are connected.
The Limit of Stars: The "Stellar Rank" (a measure of this weirdness) is much smaller than the total number of particles. It acts as a strict filter: only states with a specific, limited number of "stars" are physical and normalizable.
Perspective Matters: Whether a state is "weird" or "useful" depends on your choice of measurement basis. The authors generalized the Stellar Rank so it works for any way of measuring, not just the standard one.

In a Nutshell:
The paper tells us that to build a quantum computer with light, we shouldn't just focus on "adding more photons." Instead, we need to focus on entanglement and realize that "weirdness" is a relative concept that depends on how we choose to look at the system. The "Stellar Rank" is the tool that helps us count the true quantum resources, provided we look at them through the right lens.

1. Problem Statement

The paper addresses a fundamental gap in the understanding of non-Gaussianity in continuous-variable (CV) quantum systems.

Context: Non-Gaussianity (often characterized by the negativity of the Wigner function) is widely recognized as a necessary resource for quantum advantage in bosonic quantum information processing. The stellar rank ( $r_\star$ ) has been proposed as a hierarchical measure of non-Gaussianity, defined by the number of zeros of the Bargmann function associated with a quantum state.
The Issue: While the stellar rank is mathematically well-defined, its physical origin remains obscure. It is traditionally interpreted as the number of "displaced photons" added to a Gaussian state. However, this interpretation relies on the standard CV formalism where the phase reference is treated as an implicit, classical parameter.
The Question: What is the physical mechanism generating non-Gaussianity and stellar rank when the phase reference is treated rigorously as a quantum degree of freedom subject to photon-number superselection rules (SSR)? Furthermore, how does the transition from finite-dimensional systems (where SSR applies) to the infinite-dimensional CV limit affect the definition and interpretation of these resources?

2. Methodology

The authors employ a framework that explicitly includes a quantum phase reference mode to resolve the ambiguity of the CV limit.

SSR-Compliant Representation (SSRC): Instead of starting with a single-mode CV state, they define pure states in a two-mode system (signal mode $a$ and reference mode $b$ ) with a fixed total photon number $N$ . These states satisfy the photon-number superselection rule:
$|\psi\rangle_S = \sum_{n=0}^N c_n |n\rangle_a |N-n\rangle_b$
Majorana Polynomial Representation: They map these SSRC states to Majorana polynomials $P_N(z)$ , which characterize the state via a constellation of roots (zeros) on the complex plane (via stereographic projection). This representation is exact for finite $N$ .
The CV Limit Analysis: The authors rigorously analyze the limit $N \to \infty$ to recover the standard CV description (Bargmann-Fock space). They investigate the convergence of the Majorana polynomial $P_N(z/\sqrt{N})$ to the Bargmann function $B(z)$ .
Normalization Constraints: A critical part of their method involves analyzing the normalization condition. They demonstrate that not all limits of Majorana polynomials yield valid physical CV states; specific constraints on the growth of polynomial coefficients are required to ensure the resulting function is normalizable with respect to the Gaussian measure.
Mathematical Tools: The analysis utilizes Montel's theorem (for uniform convergence of holomorphic functions) and Hurwitz's theorem (relating the zeros of a sequence of functions to the zeros of the limit function) to establish the relationship between Majorana roots and the stellar rank.

3. Key Contributions

Physical Origin of Stellar Rank: The paper establishes that stellar rank and quadrature non-Gaussianity are direct witnesses of particle entanglement between the signal mode and the quantum phase reference.
Reinterpretation of the CV Limit: The authors prove that the CV limit is not merely a mathematical abstraction but a specific physical regime where the phase reference becomes effectively classical. They show that the CV limit is necessary and sufficient for a state to be representable as a normalizable function in Bargmann-Fock space.
Hierarchy of Roots: They clarify that the stellar rank $r_\star$ is not simply the total number of Majorana roots ( $N$ ). Instead, $r_\star$ corresponds only to a subset of roots that remain within the "CV region" ( $|z|^2 \ll \sqrt{N}$ ). The scaling bound is derived as $r_\star \leq K \ll \sqrt{N}$ .
Basis-Dependence of Quantum Advantage: The paper generalizes the concept of stellar rank to arbitrary computational bases. It demonstrates that non-Gaussianity and the associated computational advantage are relative to the choice of the computational basis (or phase space representation).

4. Key Results

Entanglement vs. Non-Gaussianity:
- A state with zero stellar rank ( $r_\star = 0$ ) corresponds to a coherent state (in the CV limit), which arises from a particle-separable SSRC state (a Fock state in a rotated basis).
- A state with nonzero stellar rank ( $r_\star \neq 0$ ) implies the underlying SSRC state is particle-entangled.
- Crucial Distinction: While $r_\star \neq 0$ implies entanglement, the converse is not true. Squeezed states (which are particle-entangled in the SSRC picture) have $r_\star = 0$ in the standard quadrature basis. Thus, stellar rank is a sufficient but not necessary witness for particle entanglement.
The "Infinity" of Zeros:
- In the standard CV formalism, the zeros of the Bargmann function for a Fock state are said to be "at infinity." The authors show this is a consequence of the roots of the Majorana polynomial lying outside the CV region ( $|z| \gg \sqrt{N}$ ) as $N \to \infty$ .
- For Schrödinger cat states, the authors show that recovering the infinite number of Bargmann roots requires a specific hierarchy of limits where the root index $k$ satisfies $k \ll N$ but $k \to \infty$ .
Generalized Stellar Rank:
- By applying unitary transformations to the phase reference, the authors show that a state Gaussian in one phase space can appear non-Gaussian in another.
- This leads to a generalized stellar rank defined relative to an arbitrary computational basis. This generalized rank serves as a witness for potential quantum advantage relative to that specific basis.

5. Significance

Unification of Resources: The work unifies the concepts of non-Gaussianity and particle entanglement. It resolves the paradox of how single-mode non-Gaussianity can exist by revealing its hidden multimode nature (entanglement with the phase reference).
Refined Resource Theory: It provides a rigorous physical foundation for the stellar hierarchy, moving beyond the heuristic "photon addition" interpretation. It clarifies that the CV limit selects a specific subset of physical states (those with bounded energy growth relative to $N$ ) that admit a consistent continuous-variable description.
Quantum Advantage: The findings suggest that the potential for quantum advantage is not an absolute property of a state but is basis-dependent. A state may be a resource for quantum computation in one encoding (basis) while being classically simulable in another.
Broader Impact: This framework extends beyond quantum optics to other bosonic platforms, such as Bose-Einstein condensates, offering a new perspective on how superselection rules constrain and define quantum resources in macroscopic systems.

In summary, the paper redefines non-Gaussianity not as an intrinsic property of a single mode, but as a manifestation of entanglement with a quantum reference frame, providing a rigorous mathematical and physical bridge between finite-dimensional superselection-compliant states and infinite-dimensional continuous-variable quantum information.