论文标题
基于MACH-ZEHNDER干涉仪的数据驱动建模基于光学矩阵乘数
Data-driven Modeling of Mach-Zehnder Interferometer-based Optical Matrix Multipliers
论文作者
论文摘要
光子综合电路正在促进光学神经网络的发展，这与电子同步物具有更快，更节能的潜力，因为光学信号非常适合实现矩阵乘法。但是，光子矩阵乘法的光子芯片的准确编程仍然是一个困难的挑战。在这里，我们描述了简单的分析模型和数据驱动的模型，用于离线矩阵乘数的离线训练。我们使用从制造的芯片中获得的实验数据训练和评估模型，该芯片具有Mach-Zehnder干涉仪网状网状，该芯片实现了3 x-3矩阵乘法。基于神经网络的模型在预测错误方面优于基于物理的模型。此外，神经网络模型还能够预测覆盖C波段多达100个频率通道的矩阵权重的光谱变化。使用神经网络模型编程用于光学矩阵乘法的芯片可提高多个机器学习任务的性能。
Photonic integrated circuits are facilitating the development of optical neural networks, which have the potential to be both faster and more energy efficient than their electronic counterparts since optical signals are especially well-suited for implementing matrix multiplications. However, accurate programming of photonic chips for optical matrix multiplication remains a difficult challenge. Here, we describe both simple analytical models and data-driven models for offline training of optical matrix multipliers. We train and evaluate the models using experimental data obtained from a fabricated chip featuring a Mach-Zehnder interferometer mesh implementing 3-by-3 matrix multiplication. The neural network-based models outperform the simple physics-based models in terms of prediction error. Furthermore, the neural network models are also able to predict the spectral variations in the matrix weights for up to 100 frequency channels covering the C-band. The use of neural network models for programming the chip for optical matrix multiplication yields increased performance on multiple machine learning tasks.
