🔢 mathematics

Quantum Behaviors Are Not Semialgebraic

This paper resolves Tsirelson's 1993 open question by proving that the set of quantum behaviors in a Bell scenario with four binary measurements per party is not semialgebraic, thereby demonstrating that it cannot be exactly characterized by any finite level of the Navascués-Pironio-Acín hierarchy or described by finite real-analytic equations.

Original authors: Minbo Gao, Zhengfeng Ji, Chenghua Liu

Published 2026-09-17

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Minbo Gao, Zhengfeng Ji, Chenghua Liu

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

In the strange world of quantum physics, particles can become linked in ways that defy our everyday experience. When two such linked particles are measured by distant observers, the results they get are not random in the usual sense; they are correlated in a pattern that no ordinary, pre-agreed plan could ever produce. This phenomenon, known as quantum nonlocality, is the engine behind the most secure forms of communication and the most powerful future computers. To understand how these machines work, scientists map out every possible pattern of results that quantum mechanics allows. They call these patterns "behaviors." For decades, researchers have wondered if the entire collection of these quantum patterns has a simple, finite mathematical description. Specifically, they asked if the rules governing these patterns could be written down using a limited number of algebraic equations and inequalities, much like how the rules of a game might be listed in a rulebook. This question, posed by the physicist Boris Tsirelson in 1993, has remained unanswered, leaving a gap in our understanding of the fundamental geometry of the quantum world.

A team of researchers has now provided a definitive answer to this long-standing question, and the result is a surprise. They proved that the set of all possible quantum behaviors cannot be described by any finite collection of polynomial equations and inequalities. In simpler terms, the shape of the quantum world is too complex to be captured by a finite rulebook of algebraic formulas. The researchers focused on a specific experimental setup where two people, each performing four different measurements with two possible outcomes, share a quantum link. By constructing a very specific, smooth curve of possible results that passes through this setup, they showed that the points where quantum mechanics actually allows a result are scattered along this curve in a way that no finite algebraic description can capture. The allowed points form a sequence that gets closer and closer to a single, simple classical point, but the gaps between them never disappear. Because any finite algebraic description would have to include a continuous stretch of the curve near that point, and because the quantum reality only allows specific, isolated points there, the description fails.

This finding rules out the possibility that the quantum set can be exactly defined by a finite number of constraints, even if we allow for extra variables or more complex combinations of equations. It also shows that the current best methods for approximating quantum behaviors, known as the Navascués–Pironio–Acín hierarchy, can never be perfect. These methods work by creating a series of increasingly accurate approximations, but the researchers demonstrated that no matter how far you go in this series, you will always include some impossible results near the edge of the quantum set. The study reveals that the boundary between what is possible and what is impossible in quantum mechanics is not a smooth, simple surface, but a jagged, infinitely detailed structure that resists being pinned down by finite formulas.

The researchers arrived at this conclusion by designing a mathematical path that connects different quantum scenarios. They created a curve where the results depend on a single changing number. Along this curve, they found that quantum mechanics only permits results at specific, discrete values of that number, which get closer and closer together as they approach a limit. At the very end of this sequence lies a result that can be explained by classical physics, without any quantum weirdness. However, the quantum points leading up to this classical limit are not a solid block; they are like a series of stepping stones that get infinitely close to the shore but never quite merge into a continuous beach. The researchers proved that any attempt to describe the quantum set with a finite number of algebraic rules would inevitably include the entire stretch of the curve leading to the shore, thereby including results that quantum mechanics forbids. This proves that the quantum set is not "semialgebraic," a technical term meaning it cannot be described by a finite set of polynomial conditions.

The implications of this discovery extend beyond just the shape of the quantum set. It shows that the complexity of quantum correlations is intrinsic and cannot be simplified away, even if we look at the system from a distance or try to approximate it. The researchers also showed that this complexity persists even when we consider the most general mathematical models of quantum mechanics, including those where the measurements are not limited to a fixed size. They demonstrated that the difficulty in describing these behaviors is not just a matter of calculation, but a fundamental geometric property. The fact that the quantum set is not semialgebraic means that there is no finite list of rules that can perfectly separate the possible quantum outcomes from the impossible ones. This has profound consequences for how we think about characterizing quantum devices and for the limits of what we can prove about them using standard mathematical tools.

The study also highlights a fascinating contrast between the full quantum behavior and a simplified version that only looks at correlations between measurements. While the simplified version can be described by finite rules, the full picture, which includes the probabilities of every possible outcome, cannot. This suggests that the extra information contained in the full probabilities is what creates the infinite complexity. The researchers showed that even though the quantum points get closer and closer to a classical point, the quantum nature of the system does not fade away smoothly; instead, it remains distinct and discrete right up to the very edge. This behavior challenges the intuition that complex quantum systems should eventually look like simple classical ones as we approach a limit.

In the end, this work settles a decades-old question by showing that the quantum world is more intricate than previously thought. It is not a shape that can be fully captured by a finite set of algebraic equations. The researchers' proof is rigorous and covers all the major models of quantum mechanics, leaving no room for doubt. Their findings suggest that to fully understand and describe quantum behaviors, we may need to move beyond finite algebraic descriptions and embrace more complex, infinite structures. This does not mean we cannot work with quantum mechanics, but it does mean that our mathematical tools for describing it have inherent limitations. The quantum set remains a rich, complex landscape that resists simple categorization, reminding us that the universe is often more subtle and detailed than our best equations can fully express.

Technical Summary: Quantum Behaviors Are Not Semialgebraic

Problem Statement
The paper addresses a longstanding question posed by Tsirelson in 1993 regarding the mathematical structure of the set of quantum behaviors in a Bell scenario. Specifically, it asks whether the set of conditional probabilities achievable by local measurements on a shared quantum state (the quantum set, $\mathcal{C}_q$ ) is semialgebraic. A set is semialgebraic if it can be described by a finite Boolean combination of polynomial equations and inequalities with real coefficients. While the local and nonsignaling sets in fixed Bell scenarios are polytopes (defined by linear inequalities), the quantum set lies between them with a richer geometry. Although finite-dimensional quantum sets are known to be semialgebraic (via the Tarski–Seidenberg theorem), it remained an open question whether this property holds when the internal Hilbert space dimension is unbounded but the Bell scenario (number of inputs/outputs) is fixed.

Methodology
The authors construct a specific counterexample in the Bell scenario with four binary measurements per party ( $4 \times 4$ inputs, $2 \times 2$ outputs). Their approach involves the following steps:

Construction of a Quadratic Curve: They define a one-parameter family of synchronous, nonsignaling behaviors, denoted $p_\alpha$ , parameterized by a real number $\alpha \in (1, 2)$ . The probabilities are defined as quadratic functions of $\alpha$ .
Spectral Analysis of Projections: They leverage the structure of synchronous correlations, where equal inputs yield equal outputs. By analyzing the commuting-operator model ( $\mathcal{C}_{qc}$ ), they show that if a behavior $p_\alpha$ is realizable, the sum of four orthogonal projections $S = \sum_{i=1}^4 A_i$ must satisfy the operator identity $S = \alpha I$ on a specific subspace.
Application of the KRS Classification: They utilize a known classification (Kruglyak, Rabanovich, and Samoilenko) of scalar sums of four orthogonal projections. This classification dictates that for $1 < \alpha < 2$ , the equation $\sum_{i=1}^4 P_i = \alpha I$ has solutions if and only if $\alpha$ belongs to the discrete sequence $\alpha_m = 2 - 2/m$ for integers $m \ge 3$ .
Realization in Finite Dimensions: They demonstrate that for every $\alpha_m$ in this sequence, a finite-dimensional quantum realization exists using a specific construction involving symmetrized projections on a Hilbert space of dimension $24m$ .
Geometric Obstruction: They analyze the intersection of the curve $p_\alpha$ with the quantum set $\mathcal{C}_q$ , its closure $\mathcal{C}_{qa}$ , and the commuting-operator set $\mathcal{C}_{qc}$ . They prove that the set of valid parameters is exactly the discrete sequence $\{\alpha_m\}$ , which accumulates at the classical point $\alpha = 2$ (corresponding to $p_2$ ).

Key Contributions and Results

Non-Semialgebraicity: The primary result is that the sets $\mathcal{C}_q$ $C_{q}$ , $\mathcal{C}_{qa}$ $C_{q a}$ , and $\mathcal{C}_{qc}$ $C_{q c}$ in the four-input binary-output scenario are not semialgebraic.
- Proof Logic: If these sets were semialgebraic, their intersection with the polynomial curve $p_\alpha$ would yield a semialgebraic subset of the real line. However, the intersection is the infinite, countable set $\{2 - 2/m\}_{m \ge 3}$ , which accumulates at 2. A semialgebraic subset of $\mathbb{R}$ is a finite union of intervals and points; it cannot be an infinite, countable set with an accumulation point unless it contains an interval. Since the sequence does not contain an interval near the limit, the sets cannot be semialgebraic.
Local Analytic Obstruction: The paper further proves that none of these sets is semianalytic at the classical behavior $p_2$ . A semianalytic set near a point is defined by finite Boolean combinations of real-analytic functions. The authors show that any such description would imply the membership condition is constant on a punctured neighborhood of $\alpha=2$ , contradicting the discrete nature of the valid parameters.
Dimensional Scaling: The construction reveals that the minimum local dimension required to exactly realize $p_{\alpha_m}$ scales as $\Theta((2-\alpha_m)^{-1})$ , growing unbounded as the behavior approaches the classical limit $p_2$ , despite $p_2$ itself being a classical local behavior.
Failure of Finite Relaxations: The results imply that no finite level of the Navascués–Pironio–Acín (NPA) hierarchy can exactly characterize these sets. Every finite NPA level $Q_\ell$ is a semialgebraic outer approximation. Consequently, for any fixed $\ell$ , there exists a $\beta_\ell < 2$ such that $Q_\ell$ contains the entire interval $(\beta_\ell, 2)$ , including uncountably many behaviors that are not in $\mathcal{C}_{qc}$ .

Significance and Claims
The paper claims to definitively answer Tsirelson's 1993 question in the negative. Its significance lies in:

Ruling out Exact Finite Descriptions: It establishes that the quantum set cannot be described by any finite system of polynomial equations and inequalities, nor by any finite semidefinite representation (SDR), regardless of the number of auxiliary variables used.
Limitations of Hierarchies: It demonstrates a fundamental limitation of the NPA hierarchy: while it converges to the quantum set in the limit, no finite level is sufficient for an exact characterization in this scenario.
Geometric Nature of the Obstruction: Unlike previous results concerning non-closure or undecidability, this obstruction is purely geometric and persists even when considering the closure of the quantum set. It applies to the commuting-operator model as well, separating the algebraic nature of the set from the analytic nature of the limit point.
Threshold Identification: The paper identifies the onset of non-semialgebraicity at exactly four inputs within the synchronous binary sector, noting that scenarios with three inputs remain semialgebraic.

The authors emphasize that these results are derived from exact mathematical proofs and do not rely on empirical data or experimental proposals. The findings constrain the search for physical principles that might characterize quantum correlations, suggesting that such principles cannot be reduced to finite polynomial conditions if they are to capture the exact quantum set.

Technical Summary: Quantum Behaviors Are Not Semialgebraic

More like this