Abstract
Increasing the layer number of on-chip photonic neural networks (PNNs) is essential to improve its model performance. However, the successive cascading of network hidden layers results in larger integrated photonic chip areas. To address this issue, we propose the optical neural ordinary differential equations (ON-ODEs) architecture that parameterizes the continuous dynamics of hidden layers with optical ODE solvers. The ON-ODE comprises the PNNs followed by the photonic integrator and optical feedback loop, which can be configured to represent residual neural networks (ResNets) and implement the function of recurrent neural networks with effectively reduced chip area occupancy. For the interference-based optoelectronic nonlinear hidden layer, the numerical experiments demonstrate that the single hidden layer ON-ODE can achieve approximately the same accuracy as the two-layer optical ResNets in image classification tasks. In addition, the ON-ODE improves the model classification accuracy for the diffraction-based all-optical linear hidden layer. The time-dependent dynamics property of ON-ODE is further applied for trajectory prediction with high accuracy.
© 2023 Optica Publishing Group
Full Article | PDF ArticleMore Like This
Jie Hou, Jianji Dong, and Xinliang Zhang
Opt. Lett. 42(12) 2310-2313 (2017)
Ying Huang, Weiping Wang, Lei Qiao, Xiaoyan Hu, and Tao Chu
Opt. Lett. 47(7) 1810-1813 (2022)
Jiahao Zhou, Pengxing Guo, Jingsi Li, Bing Lu, Luan H. K. Duong, Weigang Hou, and Lei Guo
Opt. Express 31(21) 34154-34168 (2023)