An optical neural network using less than 1 photon per multiplication

Deep learning has become a widespread tool in both science and industry. However, continued progress is hampered by the rapid growth in energy costs of ever-larger deep neural networks. Optical neural networks provide a potential means to solve the energy-cost problem faced by deep learning. Here, we experimentally demonstrate an optical neural network based on optical dot products that achieves 99% accuracy on handwritten-digit classification using ~3.1 detected photons per weight multiplication and ~90% accuracy using ~0.66 photons (~2.5 × 10−19 J of optical energy) per weight multiplication. The fundamental principle enabling our sub-photon-per-multiplication demonstration—noise reduction from the accumulation of scalar multiplications in dot-product sums—is applicable to many different optical-neural-network architectures. Our work shows that optical neural networks can achieve accurate results using extremely low optical energies.

Statistical Physical Properties of Four-Wave Mixing Optical Neural Network

Исследованы статфизические свойства оптической нейросети. Получены условия, при которых возможно обучение нейросети алгоритмом максимального правдоподобия. Исследование проведено на примере трехмерной модели Изинга, в которой последовательно добавляется дальнодействие так, что в пределе модель можно описывать теорией среднего поля. Получены аналитические оценки для критической температуры нейросети при учете взаимодействия со вторыми и третьими соседями. Данные оценки на всем интервале значений параметров взаимодействия хорошо согласуются с результатами, полученными методами Монте-Карло. Установлено, что с ростом числа положительных межсвязей величина критической температуры падает и алгоритм максимального правдоподобия может применяться практически без ограничений. The paper investigates the statistical physical properties of an optical neural network. The conditions for training a neural network by the maximum likelihood algorithm are identified. The study uses a three-dimensional Ising model, to which a long-range action is sequentially added so that in the limit the model can be described by the mean-field theory. Analytical estimates of the critical neural network temperature were obtained considering the interaction with the second and third-order neighbors. The estimates for the entire interval of the interaction parameters are in good agreement with the results obtained by Monte Carlo methods. It is found that as the number of positive interconnections increase, the critical temperature value decreases and the maximum likelihood algorithm can be applied virtually without any restrictions.

Implementation of Pruned Backpropagation Neural Network Based on Photonic Integrated Circuits

We demonstrate a pruned high-speed and energy-efficient optical backpropagation (BP) neural network. The micro-ring resonator (MRR) banks, as the core of the weight matrix operation, are used for large-scale weighted summation. We find that tuning a pruned MRR weight banks model gives an equivalent performance in training with the model of random initialization. Results show that the overall accuracy of the optical neural network on the MNIST dataset is 93.49% after pruning six-layer MRR weight banks on the condition of low insertion loss. This work is scalable to much more complex networks, such as convolutional neural networks and recurrent neural networks, and provides a potential guide for truly large-scale optical neural networks.

Research progress in optical neural networks: theory, applications and developments

With the advent of the era of big data, artificial intelligence has attracted continuous attention from all walks of life, and has been widely used in medical image analysis, molecular and material science, language recognition and other fields. As the basis of artificial intelligence, the research results of neural network are remarkable. However, due to the inherent defect that electrical signal is easily interfered and the processing speed is proportional to the energy loss, researchers have turned their attention to light, trying to build neural networks in the field of optics, making full use of the parallel processing ability of light to solve the problems of electronic neural networks. After continuous research and development, optical neural network has become the forefront of the world. Here, we mainly introduce the development of this field, summarize and compare some classical researches and algorithm theories, and look forward to the future of optical neural network.