Fully Forward Mode (FFM) is a method for training optical neural networks. It was proposed by the research team of Academician Dai Qionghai and Professor Fang Lu of Tsinghua University in 2024. The relevant paper is "Fully forward mode training for optical neural networks", which details the principle, implementation and application of the FFM method, and demonstrates its effectiveness and superior performance in training different optical systems. This research result was published in the journal Nature in 2024, marking a major breakthrough in the field of optical computing and artificial intelligence.

This method uses the symmetry of photon propagation to equate both forward and backward propagation in neural network training to the forward propagation of light, thereby developing an efficient optical neural network training method. Through FFM learning, researchers are able to train deep optical neural networks (ONNs) with millions of parameters and achieve ultra-sensitive perception and efficient all-optical processing.