论文标题
受语言建模启发的量子状态断层扫描
Quantum State Tomography Inspired by Language Modeling
论文作者
论文摘要
量子状态断层扫描是完全表征未知量子状态的基本工具。随着量子硬件大小的扩大，由于其成倍增长的复杂性，标准量子状态层析成像变得越来越具有挑战性。在这项工作中，我们通过将状态层析成像视为一种语言建模任务来提出一个可扩展的解决方案，其中未知的量子状态被视为一种未知语言，量子状态的相关性被解释为特定于该语言的语义信息，而测量结果仅是从该语言中产生的文本实例。基于语言建模的自定义变压器模型，我们证明我们的方法可以使用比最新方法更少的样品准确地重建典型的纯量子和混合量子状态。更重要的是，与现有的神经网络方法相比，我们的方法可以同时重建一类相似状态，这些神经网络方法需要为每个未知状态训练模型。
Quantum state tomography is an elementary tool to fully characterize an unknown quantum state. As the quantum hardware scales up in size, the standard quantum state tomography becomes increasingly challenging due to its exponentially growing complexity. In this work, we propose a scalable solution by considering state tomography as a language modeling task, where the unknown quantum state is treated as an unknown language, the correlation of the quantum state is interpreted as the semantic information specific to this language, and the measurement outcomes are simply the text instances generated from the language. Based on a customized transformer model from language modeling, we demonstrate that our method can accurately reconstruct prototypical pure and mixed quantum states using less samples than state-of-the-art methods. More importantly, our method can reconstruct a class of similar states simultaneously, in comparison with the existing neural network methods that need to train a model for each unknown state.
