⚛️ quantum physics

The I3322 quantum value is attained spatially but not in finite dimension

This paper proves that the quantum supremum of the $I_{3322}$ Bell functional is attained only by an infinite-dimensional spatial strategy on $\ell^2(\mathbb{Z})\otimes\ell^2(\mathbb{Z})$ and not by any finite-dimensional quantum system, thereby confirming the Pal-Vertesi conjecture, establishing the non-closure of the set of quantum correlations for this scenario, and determining its logarithmic dimension complexity.

Original authors: Seth Douglas

Published 2026-09-09

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Seth Douglas

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 quantum world, particles can become linked in a way that defies our everyday experience of distance and separation. When two such particles are measured, the result of one measurement can instantly influence the other, no matter how far apart they are. Scientists describe these connections using mathematical rules called Bell inequalities, which act like a test to see if the universe follows the strange laws of quantum mechanics or the more predictable laws of classical physics. For decades, researchers have known that to get the strongest possible quantum connections, you sometimes need to use systems that are infinitely large, rather than just a few simple particles. However, a specific and famous test known as the I3322 inequality had remained a stubborn mystery. It was suspected that this test required an infinite system to reach its maximum possible score, but no one could prove it, nor could they show exactly how an infinite system would achieve it. This uncertainty left a gap in our understanding of the very limits of quantum reality.

A new study has finally closed that gap, proving that the I3322 test cannot be won by any system made of a finite number of parts, no matter how complex or large that system is. The researcher demonstrated that to reach the absolute highest score allowed by quantum mechanics for this specific test, one must use an infinite-dimensional system, a mathematical structure that goes on forever. They did not just show that a finite system fails; they proved the existence of an infinite system that achieves the perfect score by analyzing the mathematical properties of the system's spectral measure. This finding settles a long-standing debate that began over a decade ago, confirming that for this particular scenario, the quantum world truly demands an infinite scale to reach its full potential.

The researcher behind this discovery, led by Seth Douglas, approached the problem by treating the quantum system like a chain of linked values. They started by establishing a very precise range for the maximum possible score, narrowing it down to a tiny window between two specific numbers. Within this window, they proved that any attempt to reach the top score using a finite number of dimensions would inevitably fall short. Their proof relied on a clever logical trap: they showed that if you try to force a finite system to reach the maximum, the mathematical rules governing the system would force it into a state where the score is capped at a lower value, roughly one-quarter, which is strictly less than the true maximum. This cap is a hard limit for any finite system, meaning that no matter how many particles or how much complexity you add, you can never cross that threshold to reach the true quantum peak.

Having ruled out finite systems, the researcher then turned to the question of how to actually build the infinite system that works. They proved that such a system exists by decomposing the mathematical description of the quantum state into an infinite chain of values. The key to their success was showing that the "weights" or strengths of the connections in this chain naturally fit together in a way that allows the system to be stable and real, without needing to be artificially forced. They found that the system behaves like a wave that decays as it moves away from the center, ensuring that the total energy remains manageable even though the system is infinite. This construction proved that the maximum score is not just a theoretical limit that can be approached but never touched; it is a value that can be exactly reached, but only by stepping into the realm of the infinite.

The study also addressed how quickly a finite system can get close to this perfect score. The researcher found that to get within a tiny fraction of the maximum value, the size of the system must grow logarithmically. In simpler terms, if you want to get twice as close to the perfect score, you do not need to double the size of your system; you only need to add a small, fixed amount of complexity. This means that while you can never reach the top with a finite system, you can get very close to it with a surprisingly modest increase in size. This finding provides a clear map of the cost of approximation, showing that the path to the quantum limit is efficient, even if the destination itself is unreachable for finite beings.

This work resolves a challenge that was explicitly posed by other scientists in 2010, who had suspected that finite systems were insufficient but could not prove it. The new paper confirms their suspicion and goes further by providing the exact mathematical blueprint for the infinite system that succeeds. It also clarifies the nature of the quantum world in this specific scenario, showing that the set of all possible quantum correlations is not a closed, complete shape but has a missing piece that can only be filled by an infinite structure. By proving that the I3322 inequality requires infinite dimensions, the study adds a definitive piece to the puzzle of quantum mechanics, illustrating that the universe's rules sometimes demand a scale that transcends the finite limits of any physical object we can build.

Technical Summary: The I3322 Quantum Value is Attained Spatially but Not in Finite Dimension

Problem Statement
The paper addresses the long-standing question of whether the quantum supremum of the $I_{3322}$ Bell inequality can be attained by finite-dimensional quantum systems. The $I_{3322}$ inequality involves three binary measurements per party and is a facet of the local polytope inequivalent to the CHSH inequality. Previous numerical work by Pál and Vértesi (2010) suggested that while finite-dimensional strategies converge to a value $S \approx 0.2508753845$ , no finite dimension suffices to reach the true quantum maximum, and that the maximum is only attainable in an infinite-dimensional Hilbert space. This paper provides an analytical proof of this conjecture, settling the attainment question explicitly raised by Dykema, Paulsen, and Prakash.

Methodology and Certified Inputs
The proofs rely on a reduction of the Bell functional to a spectral problem involving Jacobi matrices. The author utilizes a set of exactly certified inputs established in an accompanying public repository:

Certified Window: An exact rational enclosure for the quantum value $S$ is established as $S \in (0.2508753845015185, 0.250875388108398]$ . This relies on prior proof-carrying certificates by Mghirbi.
Equality of Models: A certified theorem establishes that the tensor-product supremum ( $\omega_{tensor}$ ) and the commuting-operator supremum ( $\omega_{commuting}$ ) are equal for this functional.
Bellman Storage: The proofs utilize a "critical Bellman storage," a concave function derived from the limit of Schur pivots, which prices the steps of a path in the label space.
Machine Checking: The combinatorial and scalar cores of the arguments are machine-checked in Lean 4. The paper distinguishes between the formalized cores (e.g., endpoint exclusion, amplitude elimination) and the analytic reductions (e.g., measure disintegration), which are provided as proof documents.

Key Contributions and Results

1. Non-Attainment in Finite Dimensions (Theorem 1)
The paper proves that no finite-dimensional quantum strategy (pure or mixed states, projective or POVM measurements) attains the value $S$ .

Mechanism: The proof analyzes the "equality module" of a hypothetical maximizer. In finite dimensions, the occupied support of the maximizer must carry two order-reversing maps (a reflection and a transport map). On a finite totally ordered set, these maps must coincide, forcing every occupied label to be paired with its reflection.
The Cap: This pairing forces the amplitude ratios to satisfy a condition that caps the Bell value at exactly $1/4$ . Since the certified value $S > 1/4$ , a finite-dimensional maximizer cannot exist.
Significance: This confirms the conjecture of Pál and Vértesi (2010) that finite dimensions are insufficient for the exact quantum maximum of $I_{3322}$ .

2. Spatial Attainment in Infinite Dimensions (Theorem 2)
The paper proves that the value $S$ is attained by a spatial strategy on the infinite-dimensional Hilbert space $\ell_2(\mathbb{Z}) \otimes \ell_2(\mathbb{Z})$ .

Mechanism: Unlike the finite case, the commuting model allows for a maximizer whose spectral measure can be disintegrated over the orbits of a group generated by the response transports. These orbits are order-isomorphic to $\mathbb{Z}$ .
Construction: By fixing an orbit, the author constructs an $\ell_2(\mathbb{Z})$ eigenvector of the associated Jacobi matrix with eigenvalue $S$ . Crucially, the normalizability of this eigenvector is inherited from the underlying probability measure rather than imposed, resolving a historical obstruction found in previous attempts.
Significance: This establishes that the quantum supremum is a maximum in the commuting-operator model (and thus in the spatial model $C_{qs}$ ), but not in the finite-dimensional tensor model ( $C_q$ ).

3. Non-Closure of $C_q(3, 3; 2, 2)$ (Corollary 3)
As a direct consequence, the set of bipartite correlations realizable with finite-dimensional strategies, $C_q(3, 3; 2, 2)$ , is not closed.

Minimality: The $(3, 3; 2, 2)$ scenario is identified as the smallest bipartite scenario (by input count) with two outcomes where non-closure is known. Previous examples required larger input sets (e.g., $(4, 4; 2, 2)$ or $(5, 5; 2, 2)$ ).

4. Dimension Complexity (Theorem 6)
The paper quantifies the cost of approximating the unattainable value $S$ using finite dimensions.

Result: The local dimension $d$ required to approximate $S$ within error $\epsilon$ scales as $D(\epsilon) = \Theta(\log(1/\epsilon))$ .
Upper Bound: Constructively, the author shows that truncating the infinite-dimensional carrier to a window of length $d$ yields an error decaying geometrically. They provide an explicit constant: $D(\epsilon) \leq 23.9010650 \log(1/\epsilon)$ for sufficiently small $\epsilon$ .
Lower Bound: A certificate chain establishes that no strategy converges faster than geometrically, ruling out polynomial or super-geometric convergence rates.

Significance and Scope
The paper claims to settle the $I_{3322}$ attainment question explicitly raised in the literature. It provides a structural dichotomy: finite-dimensional attainment is accompanied by closed (finitely recurrent) optimal carriers, while the known non-attainment phenomena are carried by infinite open chains.

The author is modest regarding the scope of their claims:

They do not claim that the specific Pál–Vértesi numerical family converges to $S$ , nor do they claim to identify the exact decimal value of $S$ beyond the certified window.
They do not claim minimality over scenarios with larger output alphabets.
The dimension complexity result is specific to the $I_{3322}$ scenario and the specific carrier constructed, though the $\Theta(\log(1/\epsilon))$ rate is presented as a general feature of this class of transport-class non-attainment.

The work distinguishes itself from concurrent independent proofs of non-attainment (e.g., by Pauwels) by providing the explicit spatial attainment construction on $\ell_2(\mathbb{Z})$ and the precise dimension-complexity rate. The results rely heavily on exact rational arithmetic and machine-checked cores to ensure the validity of the certified window and the exclusion of finite-dimensional maximizers.

Technical Summary: The I3322 Quantum Value is Attained Spatially but Not in Finite Dimension

More like this