Fig. 4: Classification results of handwritten digits using our device.
From: 120 GOPS Photonic tensor core in thin-film lithium niobate for inference and in situ training
a A block diagram of a multilayer perceptron neural network, which consists of an input layer, two hidden layers, and an output layer that provides classification outputs. b A schematic of the in situ training, a form of online training, where our IPTC handles forward propagation while the computer manages the nonlinearity function and backpropagation. c The validation accuracy as a function of epoch for in situ training (solid red line) scheme compared to that running on just a central processing unit (CPU, dashed blue line). d, e Theoretically calculated confusion matrices (purely run by the CPU) and experimental confusion matrices (run by our IPTC) using the MNIST large-scale database39. For “in situ" training, 2000 handwritten digits are used for training, and 500 digits are used for testing. Our IPTC achieves classification accuracy comparable to that achieved by the CPU.
