论文标题
平价时间对称光学神经网络
Parity-time symmetric optical neural networks
论文作者
论文摘要
最近提出，在一系列级联的Mach-Zehnder干涉仪（MZIS）上实施的光学神经网络（ONNS）最近被提议作为常规深度学习硬件的替代方法。与电子同行相比，它们有可能提供更高的能源效率和计算速度。通过利用可调相变器，可以调整每个MZIS的输出，以便模仿任意矩阵矢量乘法。这些相位变速器对于ONN的可编程性至关重要，但是它们需要大的足迹，并且相对较慢。在这里，我们提出了一种利用平等时间（PT）对称耦合器作为其构件的ONN体系结构。不用调节阶段，而是对整个数组的增益/损失对比进行调整，以作为训练网络的一种手段。我们证明，PT对称光学神经网络（PT-ONN）通过在改良的国家标准技术研究所（MNIST）数据集上执行数字识别任务来充分表达。与常规的ONN相比，PT-ONN具有可比的精度（67％vs. 71％），同时规避与变化相关的问题。我们的方法可能会导致在芯片尺度光学神经网络中快速训练的新替代途径。
Optical neural networks (ONNs), implemented on an array of cascaded Mach-Zehnder interferometers (MZIs), have recently been proposed as a possible replacement for conventional deep learning hardware. They potentially offer higher energy efficiency and computational speed when compared to their electronic counterparts. By utilizing tunable phase shifters, one can adjust the output of each of MZIs in order to enable emulation of arbitrary matrix-vector multiplication. These phase shifters are central to the programmability of ONNs, but they require large footprint and are relatively slow. Here we propose an ONN architecture that utilizes parity-time (PT) symmetric couplers as its building blocks. Instead of modulating phase, gain/loss contrasts across the array are adjusted as a means to train the network. We demonstrate that PT symmetric optical neural networks (PT-ONN) are adequately expressive by performing the digit-recognition task on the modified national institute of standard and technology (MNIST) dataset. Compared to conventional ONNs, the PT-ONN achieves a comparable accuracy (67% vs. 71%) while circumventing the problems associated with changing phase. Our approach may lead to new and alternative avenues for fast training in chip-scale optical neural networks.
