← Latest papers
⚛️ quantum physics

Verifiable Quantum Advantage via Optimized DQI Circuits

This paper presents optimized Decoded Quantum Interferometry (DQI) circuits that achieve verifiable quantum advantage with optimal asymptotic speedup for the Optimal Polynomial Intersection problem, requiring significantly fewer logical qubits and gates than factoring RSA-2048 by introducing novel quantum implementations of the Extended Euclidean Algorithm.

Original author: Tanuj Khattar, Noah Shutty, Craig Gidney, Adam Zalcman, Noureldin Yosri, Dmitri Maslov, Ryan Babbush, Stephen P. Jordan

Published 2026-11-20
📖 6 min read🧠 Deep dive

Original author: Tanuj Khattar, Noah Shutty, Craig Gidney, Adam Zalcman, Noureldin Yosri, Dmitri Maslov, Ryan Babbush, Stephen P. Jordan

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 author. For technical accuracy, refer to the original paper. Read full disclaimer

In the race to prove that quantum computers can truly outperform their classical counterparts, scientists have long faced a difficult choice. On one side are problems of immense practical value, like breaking the encryption that protects global banking. While quantum algorithms exist to solve these, the machines required are so vast and complex that they remain far beyond our current reach. On the other side are simpler, made-up puzzles designed specifically to be easy for a quantum machine to solve but hard for a classical one. These are easier to build, but they offer little proof of real-world utility and often require complex, interactive tests to verify the results. For years, the field has been stuck between these two extremes, searching for a problem that is both useful enough to matter and simple enough to build, all while providing a clear, verifiable proof that the quantum machine is doing something a classical computer cannot.

A team of researchers at Google Quantum AI has now proposed a solution that bridges this gap. They have refined a specific quantum algorithm, known as decoded quantum interferometry, and applied it to a problem called optimal polynomial intersection. This problem asks a computer to find the best possible set of coefficients for a mathematical curve that fits a scattered collection of data points within a specific, finite number system. While this sounds abstract, it is a fundamental type of optimization that appears in coding theory and cryptography. The researchers have shown that by carefully redesigning the internal machinery of the quantum computer, they can solve instances of this problem that would take a classical supercomputer longer than the age of the universe to crack, using a quantum machine that is surprisingly small and efficient.

The core of their achievement lies in how they managed the physical resources of the computer. A major bottleneck in running such algorithms has been the sheer number of "qubits," or quantum bits, required to hold information during the calculation. Previous designs for this type of problem demanded a massive amount of memory, effectively requiring a separate storage space for every single data point in the problem. The team realized that this was unnecessary. By developing a new method to generate and process information one piece at a time, and by using a clever technique to erase intermediate data without losing the quantum state, they drastically reduced the memory footprint. Instead of needing a storage space that grows with the total number of data points, their new design only requires a space that grows with the number of variables being solved for. This shift allows them to tackle large, difficult problems with a much smaller machine.

To make this work, the team had to overcome a specific hurdle in the mathematics of the algorithm: the need to reverse a complex decoding process. This process is similar to the math used in error-correcting codes, which are essential for keeping data intact when it is transmitted over noisy channels. The researchers invented new, highly efficient ways to perform this reversal on a quantum computer. They created two distinct approaches. One approach explicitly calculates the necessary mathematical values and stores them, while the other keeps the values in a compressed, implicit form that is only unpacked when needed. Both methods successfully reduced the number of physical components required, proving that the algorithm could run with far fewer resources than previously thought possible.

The researchers then analyzed their design against the best-known classical attacks. They constructed specific, difficult versions of the problem where the target data points were arranged in a way that makes them incredibly hard for classical computers to analyze. They found that for these difficult cases, a classical computer would need to perform more than 10^23 attempts to find a solution. In contrast, their optimized quantum design could solve the same problem with a circuit containing roughly 5.72 million basic logic gates and using 1,885 logical qubits. To put this in perspective, this is about one hundred times fewer gates than what is currently estimated to be needed to break a standard 2,048-bit RSA encryption key, a task often cited as the gold standard for quantum computing power.

Furthermore, the team calculated the physical requirements for building a machine capable of running this algorithm, assuming the use of a standard error-correction method that protects against noise. Their estimates suggest that a machine with approximately 800,000 physical qubits could solve a classically intractable instance of this problem in about six hours. This is a significant reduction compared to other proposals for demonstrating quantum advantage, which often require either vastly more hardware or rely on problems that are purely artificial and lack the structural depth of real-world optimization challenges.

The work also establishes a theoretical milestone. The researchers demonstrated that their method achieves the best possible speedup allowed by the laws of physics for this type of problem. As the size of the problem grows, the time required by the quantum computer grows at the slowest possible rate, while the time required by the best classical computer grows exponentially. This means that for large enough problems, the quantum advantage becomes overwhelming. The team's findings suggest that we do not need to wait for massive, million-qubit machines to see a clear, verifiable quantum advantage. Instead, by optimizing the algorithms and the way they use memory, we can reach that threshold with machines that are within the realm of near-future engineering.

This research does not just offer a new way to solve a specific math puzzle; it provides a blueprint for how to build more efficient quantum computers. The techniques they developed for managing memory and reversing complex calculations can be applied to other important algorithms, including those used for breaking encryption based on elliptic curves. By showing that a problem with real mathematical structure can be solved with fewer resources than previously thought, the team has moved the goalpost for what is considered a feasible demonstration of quantum power. They have identified a path where the quantum machine is not just a theoretical possibility, but a practical reality that can be built and tested in the coming years.

The implications extend beyond the specific numbers. The study challenges the assumption that useful, verifiable quantum advantage requires either massive hardware or contrived problems. By finding a middle ground where the problem is hard enough to matter but the solution is efficient enough to build, the researchers have provided a concrete target for hardware developers. Their work suggests that the path to a useful quantum computer may be shorter than anticipated, provided we focus on optimizing the software and the way the machine handles information, rather than just waiting for the hardware to scale up. The result is a compelling candidate for the first truly practical demonstration of a quantum computer doing something a classical computer cannot, verified by the sheer impossibility of the classical alternative.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →