Abstract:Recent advances in silicon photonic chips have made huge progress in optical computing owing to their flexibility in the reconfiguration of various tasks. Its deployment of neural networks serves as an alternative for mitigating the rapidly increased demand for computing resources in electronic platforms. However, it remains a formidable challenge to train the online programmable optical neural networks efficiently, being restricted by the difficulty in obtaining gradient information on a physical device when executing a gradient descent algorithm. Here, we experimentally demonstrate an efficient, physics-agnostic, and closed-loop protocol for training optical neural networks on chip. A gradient-free algorithm, that is, the genetic algorithm, is adopted. The protocol is on-chip implementable, physical agnostic (no need to rely on characterization and offline modeling), and gradient-free. The protocol works for various types of chip structures and is especially helpful to those that cannot be analytically decomposed and characterized. We confirm its viability using several practical tasks, including the crossbar switch and the <i>Iris</i> classification. Finally, by comparing our physics-agonistic and gradient-free method to the off-chip and gradient-based training methods, we demonstrate the robustness of our system to perturbations such as imperfect phase implementation and photodetection noise. Optical processors with gradient-free genetic algorithms have broad application potentials in pattern recognition, reinforcement learning, quantum computing, and realistic applications (such as facial recognition, natural language processing, and autonomous vehicles).The Supporting Information is available free of charge at <a class="ext-link" href="/doi/10.1021/acsphotonics.1c00035?goto=supporting-info">https://pubs.acs.org/doi/10.1021/acsphotonics.1c00035</a>.The statistical information of individuals in the final optimal generation; The results of realizing the crossbar switch on a four-mode optical processor following balanced design; The results of realizing a crossbar switch on an eight-mode optical processor following fast design; Programming a six-mode chip to realize a random T-matrix; Demonstrating the crossbar switch with multiple inputs; Blind classification with a GA-based method and numerical gradient-based method; GA-based training for handwriting digit classification; GA-based training for data set CIFAR-10 (<a class="ext-link" href="/doi/suppl/10.1021/acsphotonics.1c00035/suppl_file/ph1c00035_si_001.pdf">PDF</a>)This article has not yet been cited by other publications.

Optical neural network for unequally distributed neuron states: algorithm and application

The Realization Of Parallel Subtraction Using Lclv In Optical Neural Network

Countering Variations and Thermal Effects for Accurate Optical Neural Networks.

Optical Analog Computing Devices Designed by Deep Neural Network

Optical Implementation of Neural Learning Algorithms Based on Cross-gain Modulation in a Semiconductor Optical Amplifier

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

An Optical Image Encryption Method Using Hopfield Neural Network

Slimmed optical neural networks with multiplexed neuron sets and a corresponding backpropagation training algorithm

Spatially Varying Nanophotonic Neural Networks

Opto-Electronic Hybrid Network Based on Scattering Layers

High speed and reconfigurable optronic neural network with digital nonlinear activation

Feature Enhanced Optical Interpattern Associative Neural Network

Partitionable High-Efficiency Multilayer Diffractive Optical Neural Network

Hybrid training of optical neural networks

Transferable polychromatic optical encoder for neural networks

Advances and Challenges of Optical Neural Networks

Event-driven adaptive optical neural network

1-bit Quantized On-chip Hybrid Diffraction Neural Network Enabled by Authentic All-optical Fully-connected Architecture

Efficient On-Chip Training of Optical Neural Networks Using Genetic Algorithm

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

An optical neural chip for implementing complex-valued neural network