⚛️ quantum physics

Quantum Algorithms for OPI Variants Beyond Locality and Classical Decodability

This paper extends Regev's quantum reduction framework for Optimal Polynomial Intersection (OPI) variants by introducing two novel contributions: a quantum decoder for solving linear constraints over codes with a "two-fold multiplication property" and a classical decoding approach for "histogram-local" constraints, both of which overcome previous limitations regarding classical decodability and coordinate-wise locality.

Original authors: Seyoon Ragavan, Noah Shutty

Published 2026-10-02

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Seyoon Ragavan, Noah Shutty

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 quiet, high-stakes world of cryptography, researchers often play a game of cat and mouse with mathematical structures called codes. These codes are like intricate grids of numbers used to protect information, and a central challenge is finding a specific path through the grid that satisfies a complex set of rules. For decades, the most powerful tools for solving these puzzles have been classical computers, which follow step-by-step instructions. However, a new frontier has emerged with quantum computers, machines that use the strange laws of physics to explore many possibilities at once. A key technique in this field, known as Regev's reduction, acts as a bridge, turning the difficult task of finding a valid path into a problem of decoding a noisy signal. Until now, this bridge has only been usable when the rules were simple and local—meaning each position in the grid had to follow its own independent restriction—and when a fast, standard way existed to decode the signal. If either of these conditions failed, the quantum advantage vanished, and the problem remained stuck in the realm of classical difficulty.

Two researchers, Seyoon Ragavan and Noah Shutty, have now pushed past these two restrictions, showing that quantum computers can solve these grid puzzles even when the rules are more complex and the decoding methods are more difficult. Their work, published in October 2026, demonstrates two distinct ways to break the old barriers. In the first approach, they tackle a scenario where the grid is defined by a specific type of mathematical structure called a Reed-Muller code, which is based on polynomials. In this setting, the usual method of decoding fails because the noise is too heavy for classical tools to handle. The researchers designed a new quantum decoder that exploits a hidden algebraic property: when you multiply pairs of valid grid patterns together, the result is surprisingly simple and confined to a small space. By using this "two-fold multiplication" property, their quantum algorithm can find a solution with no zero entries in a regime where the best-known classical algorithms simply cannot operate. They also discovered that a slightly stronger property, involving the multiplication of three patterns, allows for a fast classical solution, but this leaves a specific middle ground where only the quantum method works.

The second breakthrough addresses a different limitation: the nature of the rules themselves. Previously, the rules had to be local, applying to each cell of the grid independently. The researchers expanded this to include "histogram-local" constraints, which are global rules about how often each symbol can appear across the entire grid. For example, a rule might state that the number '7' can appear at most three times, while the number '8' must appear exactly twice, without caring which specific cells hold those numbers. This creates a massive, interconnected web of dependencies that makes the problem much harder for classical computers. The researchers showed that if the grid is built from Reed-Solomon codes, a quantum computer can still find a solution efficiently. They proved that even if a classical computer has unlimited time and can ask questions to a random oracle—a theoretical black box that provides random answers—it will almost certainly fail to find a solution that satisfies these global frequency rules. In contrast, the quantum algorithm succeeds with a constant probability, demonstrating a clear separation between what is possible for quantum machines and what is possible for classical ones.

The significance of this work lies in its ability to expand the territory where quantum computers offer a genuine advantage. By removing the requirement for simple, local rules and by bypassing the need for efficient classical decoders, the researchers have identified new, harder problems that are still solvable by quantum methods. They did not just suggest these possibilities; they provided concrete algorithms and rigorous proofs that these methods work for specific families of codes. In one instance, they showed that a quantum algorithm could find a solution for a grid with a specific number of variables and constraints where classical methods are known to fail. In another, they proved that adding global frequency constraints to a problem makes it exponentially harder for classical computers, even if the problem remains easy for quantum ones. This suggests that the power of quantum computing in cryptography is more robust and versatile than previously thought, capable of navigating complex, global landscapes that were once considered impenetrable.

The researchers also explored the boundaries of their own findings, carefully distinguishing between what is proven and what remains an open question. They showed that while their quantum decoder works for the two-fold multiplication property, a classical algorithm can solve the same problem if a stronger three-fold property is present. This leaves a specific, intermediate range of parameters where the quantum advantage is most likely to be found, a region where the classical algorithms known today are insufficient. They did not claim to have solved the problem for every possible case, but rather to have identified and solved specific, challenging variants that were previously out of reach. Their work stands as a testament to the evolving landscape of quantum algorithms, where the focus is shifting from simple, isolated constraints to complex, global structures, and where the quantum computer's ability to navigate these structures is becoming increasingly clear.

Technical Summary: Quantum Algorithms for OPI Variants Beyond Locality and Classical Decodability

1. Problem Definition

The paper addresses the Code Intersection Problem (CIP): given a linear code $C \subseteq \mathbb{F}_q^m$ and a set $A \subseteq \mathbb{F}_q^m$ defined by nonlinear constraints, find a vector $x \in C \cap A$ . The authors focus on two specific regimes where existing quantum algorithms based on Regev's reduction (also known as decoded quantum interferometry) have been limited:

Classical Decodability: Previous successful applications relied on the ability to classically decode the dual code $C^\perp$ after measuring a noisy quantum state. This restricted the scope to codes with efficient classical decoders (e.g., low-degree polynomial codes in specific regimes).
Coordinate-wise Constraints: Previous applications required $A$ to be a Cartesian product $A_1 \times \dots \times A_m$ , where constraints apply independently to each coordinate. This excluded global constraints on symbol frequencies (histograms).

The paper seeks to overcome these restrictions separately to identify new candidates for quantum advantage and establish quantum-classical separations.

2. Methodology

The core framework utilized is Regev's reduction, which reduces the CIP to a Quantum Decoding Problem (QDP) for the dual code $C^\perp$ . Given a state $|\psi\rangle$ supported on $A$ , the reduction requires recovering a random codeword $c \in C^\perp$ from the state $\sum_c X_c \text{QFT}^\dagger |\psi\rangle$ .

The paper develops two distinct contributions to extend this framework:

Contribution 1: Beyond Classical Decodability (Quantum Decoding)

Goal: Solve the problem of finding $y \in (\mathbb{F}_q \setminus \{0\})^m$ such that $By = 0$ (where rows of $B$ span $C^\perp$ ) without assuming an efficient classical decoder for $C^\perp$ .
Methodology:
- The authors adapt the template of Chen, Liu, and Zhandry (CLZ22) but replace the reliance on full quadratic monomial spaces with a "two-fold multiplication property."
- They define the space $W = \text{span}\{u \odot v : u, v \in C^\perp\}$ , where $\odot$ is coordinate-wise multiplication. If $\dim(W) = s_2 \ll m$ , the number of auxiliary variables required for relinearization is reduced.
- Quantum Step: Instead of measuring in the standard basis (which yields noisy codewords), they use unambiguous measurements on each coordinate to obtain a pair $\{c_i, c'_i\}$ containing the true symbol.
- Linear Algebra Step: The pairs yield quadratic equations $(c_i - a_i)(c_i - a'_i) = 0$ . By lifting these to the space $W$ , the system becomes linear in $n + s_2$ variables.
- Instantiation: They apply this to Reed–Muller (RM) codes punctured at random points. For RM codes, the product of two degree- $r$ polynomials has degree at most $2r$ . This allows solving instances with $m \le n^{2-\Omega(1)}$ , a regime where efficient classical decoders for the dual RM code are not known.

Contribution 2: Beyond Product Constraints (Histogram-Local Constraints)

Goal: Solve CIP where $A$ is defined by histogram-local constraints (global constraints on the frequency of symbols), potentially after applying arbitrary permutations $\pi_i$ to each coordinate.
Methodology:
- They introduce the concept of replacement stability. They analyze the probability $p_{stay}$ that a uniform element of $A$ remains in $A$ after replacing a single coordinate with a random symbol.
- Fourier Analysis: They prove that the expected relative Hamming weight of the Fourier transform of the uniform state $|A\rangle$ is exactly $1 - p_{stay}$ .
- Decoding Strategy: If $1 - p_{stay}$ is sufficiently small (specifically, less than the list-decoding radius of $C^\perp$ ), a classical list decoder can succeed with constant probability.
- Stability Calculation: Using tilted Poisson distributions and a local central limit theorem, they show that for broad families of constraints (e.g., "max-load" constraints where symbols appear at most $T$ times), $p_{stay}$ is bounded away from zero, ensuring sufficient Fourier mass.
- Oracle Separation: In the random-oracle model, they prove that while quantum algorithms can satisfy these constraints with constant probability, classical algorithms making polynomial queries fail with exponentially small probability. This relies on the code's list-recoverability and the resistance of the histogram constraints to random completions.

3. Key Results

Result 1: Quantum Advantage in Subquadratic Regimes for RM Codes

Theorem 1.1 (Informal): For a matrix $B$ with rows spanning a code $C^\perp$ , if the dimension of the two-fold product space is $s_2$ , a quantum algorithm finds a full-support kernel vector if $\text{dist}(C^\perp) \log(q-1) \ge 2n + s_2 + 2$ .
Application: For randomly punctured Reed–Muller codes, this yields a polynomial-time quantum algorithm for $m \le n^{2-\Omega(1)}$ .
Classical Comparison: The authors also provide a classical algorithm using a "three-fold multiplication property" (requiring $\text{dist}(C^\perp) \ge n + s_3$ ).
Gap: There exist parameter regimes (specifically for RM codes) where the quantum condition (involving $s_2$ ) is satisfied, but the classical condition (involving $s_3$ ) is not, and where $m < n^2$ . These regimes represent candidates for quantum advantage, as prior classical algorithms (ISS12, IS15, etc.) only cover the $m = \Omega(n^2)$ regime.

Result 2: Quantum-Classical Separation for Histogram Constraints

Theorem 1.2 (Informal): For Reed–Solomon codes with rate $7/8$ $7/8$ and specific histogram constraints (e.g., partitioning symbols into classes with allowed counts $\{0\}$ ${0}$ , $\{1, 2\}$ ${1, 2}$ , and $\{3, 4\}$ ${3, 4}$ ):
- Plain Model: A quantum algorithm finds a satisfying codeword with inverse-polynomial probability.
- Random Oracle Model: A quantum algorithm succeeds with constant probability, while any classical algorithm making polynomial queries succeeds with exponentially small probability.
Significance: This demonstrates that adding global nonlinear constraints (histograms) to the Optimal Polynomial Intersection (OPI) problem preserves quantum easiness while compounding classical hardness.

4. Claims of Significance

The paper positions its work as pushing the boundaries of Regev's reduction in two distinct directions:

Breaking the Classical Decodability Barrier: The authors argue that it is "surprising and unnatural" that quantum decoders have not been utilized to solve problems where classical decoders are unknown. By leveraging the algebraic structure of product spaces (two-fold multiplication), they demonstrate that quantum algorithms can solve code intersection problems in regimes where the dual code lacks a known efficient classical decoder. This provides a concrete candidate for quantum advantage that does not rely on the "dequantized" success of previous works.
Expanding the Scope of Nonlinear Constraints: The work extends the applicability of Regev's reduction beyond coordinate-wise (local) constraints to global histogram constraints. By establishing a link between replacement stability and Fourier sparsity, the authors show that quantum algorithms can handle complex global constraints that are classically hard.
Oracle Separations: The paper provides a rigorous proof of quantum-classical separation relative to a random oracle for these new constraint types, reinforcing the potential for quantum advantage in code intersection problems beyond the specific OPI instances previously studied.

5. Modesty and Open Questions

The authors are explicit about the limitations and open nature of their results:

Dequantization: They do not claim that their quantum algorithms for Reed–Muller codes are classically hard. They explicitly state that the "obvious open question" is whether these results can be dequantized. They note that their classical algorithm using three-fold multiplication leaves intermediate regimes as candidates for quantum advantage, but a classical algorithm using only two-fold multiplication remains an open challenge.
Worst-Case vs. Average-Case: The histogram constraint results in the plain model rely on averaging over random permutations to achieve inverse-polynomial success. The authors hope future work can extend these to worst-case guarantees.
Combining Contributions: They acknowledge the technical difficulty of combining a quantum decoder with global nonlinear constraints, as the analysis of their quantum decoder relies on independent single-coordinate states, which are disrupted by global entanglement.

In summary, the paper provides a theoretical framework for extending quantum code intersection algorithms to regimes where classical decoding is unknown and to problems with global constraints, offering specific parameter regimes and oracle separations as evidence of potential quantum advantage.