Back-Propagation-Free Photonic Deep Learning with Random Nonlinear Projection
M. Nakajima, Katsuma Inoue, Kenji Tanaka, Yasuo Kuniyoshi, Toshikazu Hashimoto, Kohei Nakajima · 2023
We demonstrate back-propagation-free photonic deep learning by extending a biologically inspired training called direct feedback alignment. The potential for accelerated computation with competitive performance was confirmed through numerical simulation and optoelectric hardware implementation.