用于赊销风险分类的变分量子本征求解器
Variational Quantum Eigensolver for Classification in Credit Sales Risk
AI总结：
本文提出基于 VQE 与 SWAP-Test 的量子分类电路，将客户赊销数据编码为量子态并识别信用风险，具有低量子比特开销且可在 NISQ 设备上运行。
摘要：
数据分类任务在众多科学领域中得到广泛应用，并且可以通过不同的已知方法（例如神经网络）来实现。然而，在本工作中，我们利用量子计算来解决该问题。我们考虑了一种基于变分量子本征求解器（VQE）和所谓 SWAP-Test 的量子电路，这使我们能够解决与赊销相关的分类问题。更具体地说，我们处理的是一个决策问题，即根据所选决策变量（例如产生的营业额、合作历史）的值来确定客户的可靠性。经典数据样本被转换为归一化量子态。在此操作之后，样本可以由量子门电路处理。VQE 方法允许训练量子电路的参数（所谓的 ansatz），以为每个类别输出模式态。在所使用的数据集中，可以观察到两个类别——低信用风险和高信用风险案例。然而，VQE 电路区分的类别多于两个（引入了更细致的案例），并且最终结果借助上述 SWAP-Test 获得。所阐述的解决方案结构紧凑，并且只需要对数增长数量的量子比特（这归因于量子寄存器的指数容量）。由于所提出的量子电路复杂度较低，可以在当前可用的量子计算机上执行实验，包括含噪声中等规模量子（NISQ）设备。这类设备尽管存在噪声，仍能够解决本文所分析的任务。所有计算、模拟、绘图和比较均在 Python 语言环境中实现并开展。每个量子分类示例的源代码均可在源代码仓库中找到。
英文摘要：
The data classification task is broadly utilized in numerous fields of science and it may be realized by different known approaches (e.g. neural networks). However, in this work, quantum computations were harnessed to solve the problem. We take into consideration a quantum circuit which is based on the Variational Quantum Eigensolver (VQE) and so-called SWAP-Test what allows us to solve a classification problem connected with credit sales. More specifically, we cope with a decision problem of determining customer's reliability based on values of selected decision variables (e.g. generated turnover, history of cooperation). The classical data samples are converted into normalized quantum states. After this operation, samples may be processed by a circuit of quantum gates. The VQE approach allows training the parameters of a quantum circuit (so-called ansatz) to output pattern-states for each class. In the utilized data set, two classes may be observed -- cases with low and high credit risk. However, the VQE circuit differentiates more classes than two (introduces more detailed cases) and the final results are obtained with the use of aforementioned SWAP-Test. The elaborated solution is compact and requires only logarithmically increasing number of qubits (due to the exponential capacity of quantum registers). Because of the low complexity of the presented quantum circuit, it is possible to perform experiments on currently available quantum computers, including the Noisy Intermediate-Scale Quantum (NISQ) devices. This type of devices, despite the presence of noise, is capable of solving the task analyzed in this work. All calculations, simulations, plots, and comparisons were implemented and conduced in the Python language environment. Source codes for each example of quantum classification can be found in the source code repository.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
