近日，日本庆应义塾大学Shu Kanno团队研究了基于张量的大规模论证量子相位差估计。2025年7月24日出版的《美国科学院院刊》杂志发表了这项成果。

研究团队开发了一种利用量子相位差估计（QPDE）方案的能量计算算法和一种基于张量网络的统一压缩方法，用于制备叠加态和时间演化门。该算法在有效实现的同时，还能成倍地降低去极化噪声的影响。研究组在IBM超导器件主题电路上演示了一维Hubbard模型的能隙计算，该电路高达32个系统（plthem one-ancilla）量子比特，比之前的量子相位估计（QPE）演示增加了五倍，在7242控制的z门水平的标准编译，利用Q-CTRL错误抑制模块。此外，研究组提出了一种以空间轨道定位和索引排序为主题的分子执行技术，验证了高达21量子位的线性多烯模拟。由于QPDE可以处理与QPE相同的目标，因此他们的算法代表了实际设备上量子计算的飞跃。

附：英文原文

Title: Tensor-based quantum phase difference estimation for large-scale demonstration

Author: Kanno, Shu, Sugisaki, Kenji, Nakamura, Hajime, Yamauchi, Hiroshi, Sakuma, Rei, Kobayashi, Takao, Gao, Qi, Yamamoto, Naoki

Issue&Volume: 2025-7-24

Abstract: We develop an energy calculation algorithm leveraging quantum phase difference estimation (QPDE) scheme and a tensor-network-based unitary compression method in the preparation of superposition states and time-evolution gates. Alongside its efficient implementation, this algorithm reduces depolarization noise affections exponentially. We demonstrated energy gap calculations for one-dimensional Hubbard models on IBM superconducting devices using circuits up to 32-system (plus one-ancilla) qubits, a five-fold increase over previous Quantum phase estimation (QPE) demonstrations, at the 7242 controlled-Z gate level of standard transpilation, utilizing a Q-CTRL error suppression module. Additionally, we propose a technique toward molecular executions using spatial orbital localization and index sorting, verified linear polyene simulations up to 21 qubits. Since QPDE can handle the same objectives as QPE, our algorithm represents a leap forward in quantum computing on real devices.

DOI: 10.1073/pnas.2425026122

Source: https://www.pnas.org/doi/abs/10.1073/pnas.2425026122