中心硕士生刘滨硕的工作——Low-threshold all-optical nonlinear activation function based on injection locking in distributed feedback laser diode（基于分布式反馈激光器注入锁定的低阈值全光非线性激活函数）的相关成果近期被Optics Letters期刊接收发表，该工作得到了国家重点研发计划（2019YFB2203700）、国家自然科学基金（T2225023, 62205203）的部分资助。在实验系统中，我们演示了一种基于分布式反馈激光器注入锁定效应的全光非线性激活单元。通过优化注入锁定条件，我们利用该单元中的非线性载流子动力学产生一种低阈值非线性激活函数。该单元可以在-15.86dBm的低阈值和1GHz的高速下工作，与现有的光学非线性激活单元相比具有竞争力。我们将该单元应用于求解二阶常微分方程的神经网络任务，拟合误差低至0.0034，验证了该光学非线性激活单元的可行性。由于大规模扇出光子神经网络将显著降低一个通道中的光功率，我们的低阈值非线性激活单元方案适合于大规模高通量光子神经网络的实现。
摘要: We experimentally demonstrate an all-optical nonlinear activation unit based on injection-locking effect of distributed feedback laser diodes (DFB-LDs). The nonlinear carrier dynamics in the unit generates a low-threshold nonlinear activation function with optimized operating conditions. The unit can operate at a low threshold of -15.86 dBm and a high speed of 1 GHz, making it competitive among existing optical nonlinear activation approaches. We apply the unit to a neural network task of solving the second-order ordinary differential equation. The fitting error is as low as 0.0034, verifying the feasibility of our optical nonlinear activation approach. Given that the large-scale fan-out of optical neural networks (ONNs) will significantly reduce optical power in one channel, our low-threshold scheme is suitable for the development of high-throughput ONNs.