Tilted-Mode All-Optical Diffractive Deep Neural Networks

Diffractive deep neural networks (D²NNs) typically adopt a densely cascaded arrangement of diffractive masks, leading to multiple reflections of diffracted light between adjacent masks, thereby affecting the network's inference capability. It is challenging to fully simulate this multiple-reflection phenomenon. To eliminate this phenomenon, we designed tilted-mode all-optical diffractive deep neural networks (T-D²NNs) and proposed a theoretical model for diffraction propagation in the tilted mode. Simulation results indicate that T-D²NNs address the performance degradation caused by interlayer reflections in D²NNs constructed with high-index diffractive masks. In classification tasks, T-D²NNs achieve better classification results compared to D²NNs that consider interlayer reflections.