Deep nonlinear optical neural networks using physics-aware training
Logan G. Wright, Tatsuhiro Onodera, Martin M. Stein, Tianyu Wang, Darren T. Schachter, Zoey Hu, Peter L. McMahon · Conference on Lasers and Electro-Optics · 2021
We experimentally demonstrate deep nonlinear optical neural networks using a universal algorithm for backpropagating through arbitrary physical input-output transformations. Ultrafast second harmonic generation and other diverse processes are trained to perform image and audio classification.