用机器学习建模光干涉仪，无需解析解即可精准调控。
Leveraging machine learning features for linear optical interferometer control
- 基于监督学习构建干涉仪模型，适配具体硬件架构。
- 通过训练数据反推相位配置，实现目标幺正变换。
- 突破解析解限制，适合新型光路架构探索。
我们提出一种算法，可构建可重构光干涉仪的模型，不受特定架构约束。在对干涉仪光学模式进行幺正变换编程时，若能获得相位偏移与幺正矩阵间的解析关系，则采用解析方法；否则使用优化流程。该算法采用监督学习方式，将干涉仪模型与被研究器件的训练数据对齐。随后通过简单优化过程，利用训练好的模型确定具有特定架构的干涉仪的相位偏移，以实现所需的幺正变换。该方法无需精确解析解即可有效调制干涉仪，为探索新型干涉电路架构开辟了道路。
原文摘要 · Abstract (English)
We have developed an algorithm that constructs a model of a reconfigurable optical interferometer, independent of specific architectural constraints. The programming of unitary transformations on the interferometer's optical modes relies on either an analytical method for deriving the unitary matrix from a set of phase shifts or an optimization routine when such decomposition is not available. Our algorithm employs a supervised learning approach, aligning the interferometer model with a training set derived from the device being studied. A straightforward optimization procedure leverages this trained model to determine the phase shifts of the interferometer with a specific architecture, obtaining the required unitary transformation. This approach enables the effective tuning of interferometers without requiring a precise analytical solution, paving the way for the exploration of new interferometric circuit architectures.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。