麻豆淫院

Quantum computers, systems that perform computations leveraging quantum mechanical effects, could outperform classical computers in some optimization and information processing tasks. As these systems are highly influenced by noise, however, they need to integrate strategies that will minimize the errors they produce.

One proposed solution for enabling fault-tolerant quantum computing across a wide range of operations is known as magic state distillation. This approach consists of preparing special quantum states (i.e., magic states) that can then be used to perform a universal set of operations. This allows the construction of a universal quantum computer鈥攁 device that can reliably perform all operations necessary for implementing any quantum algorithm.

Yet while magic state distillation techniques can achieve good results, they typically consume large numbers of error-protected qubits and need to perform many rounds of error correction. This has so far limited their potential for real-world applications.

Researchers at Alice & Bob and PSL University CNRS, Inria, recently introduced unfolded distillation, an alternative protocol that reduces the resources required to prepare magic states. The newly developed protocol, outlined in a paper on the arXiv preprint server, was found to enable the preparation of magic states with fewer qubits and fewer operations, provided that the qubits exhibit a noise bias.

"The main objective was to determine to what extent we could reduce the cost of magic state preparation for a biased-noise qubit architecture," Diego Ruiz, author of the paper and Ph.D. student at Alice & Bob, told 麻豆淫院. "Magic states are a crucial part of quantum computing, as they enable us to implement all the gates needed to perform any possible quantum algorithm. While it is known that biased-noise qubits, such as cat qubits, could greatly reduce the cost of quantum memory, their impact on quantum computation with magic states was less clear."

In many earlier studies, the preparation of magic states proved to be the more resource-intensive aspect of quantum computation. Biased-noise qubits are designed to improve hardware efficiency by suppressing one type of quantum error鈥攂it flips, in the case of Alice & Bob's cat qubits. The key objective of the recent work by Ruiz and his colleagues was to design a scheme that harnesses a quantum system's noise bias to reduce the overhead of magic state preparation, and that could also be applied in practical settings employing 2D qubit layouts and with realistic error rates.

"We introduce a method called unfolded distillation that reduces the cost of magic state preparation both in space and time," explained Ruiz. "Magic state distillation is one of the primary ways to prepare magic states, but because it operates at the logical level, it is very costly. Our construction runs distillation at the physical level instead, taking advantage of the fact that biased-noise qubits remove the need for certain error checks."

The error-correcting code employed by most magic state distillation techniques is the so-called Reed-Muller code. This is a code that was originally designed for classical computing systems but was later adapted for quantum error correction.

"The Reed-Muller code requires a 3D architecture for qubits without noise bias, which is highly challenging to implement in practice," said Ruiz.

"This code can be unfolded in 2D when working with qubits with a noise bias, as Alice & Bob's do. For instance, this unfolding lets us prepare high-fidelity magic states with only about 53 qubits and about 5.5 rounds of error correction at very high noise bias."

When the researchers simulated their proposed protocol, they found that it attained a logical error rate of about 10鈦烩伔 using over an order of magnitude fewer qubit cycles than state-of-the-art schemes for biased and unbiased noise qubits. Notably, the protocol also works with two-qubit gates arranged in a 2D configuration, meaning that it is compatible with current hardware architectures, such as those based on superconducting qubits and cat qubits.

In the future, the unfolded distillation protocol could help to reduce the size and cost of the systems required to prepare magic states. This could, in turn, help to improve the performance and reliability of quantum computing systems.

"Our protocol moves us closer to hardware-efficient, fault-tolerant quantum computation for platforms like superconducting cat qubits," added Ruiz. "Several directions are interesting to pursue next. One is to push the scheme to even higher fidelities by increasing code distance, so it can be used standalone without concatenation with the usual distillation protocols. Another could be to prepare magic states for implementing the Toffoli gate directly, which is very useful in some algorithms."

漏 2025 Science X Network