Abstract:Ever-growing deep-learning technologies are making revolutionary changes for modern life.However,conventional computing architectures are designed to process sequential and digital programs but are burdened with performing massive parallel and adaptive deep-learning applications.Photonic integrated circuits provide an efficient approach to mitigate bandwidth limitations and the power-wall brought on by its electronic counterparts,showing great potential in ultrafast and energy-free high-performance computation.Here,we propose an optical computing ar-chitecture enabled by on-chip diffraction to implement convolutional acceleration,termed"optical convolution unit"(OCU).We demonstrate that any real-valued convolution kernels can be exploited by the OCU with a promi-nent computational throughput boosting via the concept of structral reparameterization.With the OCU as the fundamental unit,we build an optical convolutional neural network(oCNN)to implement two popular deep learn-ing tasks:classification and regression.For classification,Fashion Modified National Institute of Standards and Technology(Fashion-MNIST)and Canadian Institute for Advanced Research(CIFAR-4)data sets are tested with accuracies of 91.63％and 86.25％,respectively.For regression,we build an optical denoising convolutional neural network to handle Gaussian noise in gray-scale images with noise level σ=10,15,and 20,resulting in clean images with an average peak signal-to-noise ratio(PSNR)of 31.70,29.39,and 27.72 dB,respectively.The proposed OCU presents remarkable performance of low energy consumption and high information density due to its fully passive nature and compact footprint,providing a parallel while lightweight solution for future compute-in-memory architecture to handle high dimensional tensors in deep learning.

Solving Computer Vision Tasks with Diffractive Neural Networks

Design of task-specific optical systems using broadband diffractive neural networks

In Situ Optical Backpropagation Training of Diffractive Optical Neural Networks

<i>In situ</i> optical backpropagation training of diffractive optical neural networks

Scientific Computing with Diffractive Optical Neural Networks

On-chip 4F-System Based on Concave Mirrors for Optical Neural Networks

Scale-, shift- and rotation-invariant diffractive optical networks

High Throughput Multi-Channel Parallelized Diffraction Convolutional Neural Network Accelerator

Advances and Progress of Diffractive Deep Neural Networks

Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit

Pluggable Multitask Diffractive Neural Networks Based on Cascaded Metasurfaces

Space-efficient optical computing with an integrated chip diffractive neural network

Time-lapse image classification using a diffractive neural network

Sophisticated Deep Learning with On-Chip Optical Diffractive Tensor Processing

Optical backpropagation training method and its applications

High-frame-rate Reconfigurable Diffractive Neural Network Based on Superpixels.

Integration of Programmable Diffraction with Digital Neural Networks

Decision-making and control with diffractive optical networks

Diffractive Deep Neural Networks at Visible Wavelengths

Metasurface-enabled on-chip multiplexed diffractive neural networks in the visible

Deep Learning-designed Diffractive Neural Networks