Opt Lett. 2022 Apr 1;47(7):1810-1813. doi: 10.1364/OL.451287.
We experimentally demonstrate two types of programmable, low-threshold, optically controlled nonlinear activation functions, which are challenging to realize in photonic neural networks (PNNs). These devices rely on on-chip integrated Ge-Si photoelectric detectors and silicon electro-optical switches, and they generate rectified linear unit (ReLU) or sigmoid functions with arbitrary slopes without additional electrical processing. Both devices function at an extremely low threshold of 0.2 mW. The embedding of these nonlinear activation functions into convolutional neural networks facilitates the attainment of high inference accuracies of up to 95% when applied to Modified National Institute of Standards and Technology (MNIST) handwritten digit-classification tasks. The devices are suitable for low-power PNNs with an arbitrary number of propagation layers in photonic-computing chips.
我们实验演示了两种可编程、低阈值、光控非线性激活函数，这在光子神经网络（PNN）中很难实现。这些器件依赖于片上集成的 Ge-Si 光电探测器和硅电光开关，它们在无需额外电处理的情况下产生具有任意斜率的整流线性单元（ReLU）或 sigmoid 函数。这两种器件的工作阈值极低，仅为 0.2mW。当将这些非线性激活函数嵌入卷积神经网络中时，应用于修改后的国家标准与技术研究所（MNIST）手写数字分类任务时，可实现高达 95%的高推断精度。这些器件适用于具有任意数量传播层的低功耗 PNN 光子计算芯片。
