This article analyzes the performance and energy efficiency of Netcast, a recently proposed opticalneural-network architecture designed for edge computing. Netcast performs deep neural network inference by dividing t...
详细信息
This article analyzes the performance and energy efficiency of Netcast, a recently proposed opticalneural-network architecture designed for edge computing. Netcast performs deep neural network inference by dividing the computational task into two steps, which are split between the (cloud) server and (edge) client: (1) the server employs a wavelength-multiplexed modulator array to encode the network's weights onto an optical signal in an analog time-wavelength basis, and (2) the client obtains the desired matrix-vector product through modulation and time-integrated detection. The simultaneous use of wavelength multiplexing, broadband modulation, and integration detection allows large neuralnetworks to be run at the client by effectively pushing the energy and memory requirements back to the server. The performance and energy efficiency are fundamentally limited by crosstalk and detector noise, respectively. We derive analytic expressions for these limits and perform numerical simulations to verify these bounds.
Graph-based neuralnetworks have promising perspectives but are limited by electronic bottlenecks. Our work explores the advantages of opticalneuralnetworks in the graph domain. We propose an optical graph neural ne...
详细信息
Graph-based neuralnetworks have promising perspectives but are limited by electronic bottlenecks. Our work explores the advantages of opticalneuralnetworks in the graph domain. We propose an optical graph neural network (OGNN) based on inverse-designed optical processing units (OPUs) to classify graphs with optics. The OPUs, combined with two types of optical components, can perform multiply-accumulate, matrix-vector multi-plication, and matrix-matrix multiplication operations. The proposed OGNN can classify typical non-Euclidean MiniGCDataset graphs and successfully predict 1000 test graphs with 100% accuracy. The OPU-formed optical- electrical graph attention network is also scalable to handle more complex graph data, such as the Cora dataset, with 89.0% accuracy.(c) 2022 Optica Publishing Group
In this work we discuss a diffractive opticalneural network approach for recognizing the mode of OAM waves and their superposition. Experimentally, diffractive neuralnetworks are fabricated through an imprinting tec...
详细信息
ISBN:
(纸本)9798350370331;9798350370324
In this work we discuss a diffractive opticalneural network approach for recognizing the mode of OAM waves and their superposition. Experimentally, diffractive neuralnetworks are fabricated through an imprinting technique with low loss parowax material. We also show that the diffractive networks can enable mathematical operations through the topological charges of the superposed OAM waves, being capable of displaying these results in a digit format across different operation planes. The approaches herein are general and could be used to arbitrarily manipulate multiple superposed OAM states, which can enable a myriad of potential applications in the next generation of terahertz communications systems.
We deploy reconfigurable diffractive opticalneuralnetworks for multiple scientific computing applications, including guiding quantum material synthesis, predicting properties of materials, biomolecules, and nanophot...
详细信息
We deploy reconfigurable diffractive opticalneuralnetworks for multiple scientific computing applications, including guiding quantum material synthesis, predicting properties of materials, biomolecules, and nanophot...
详细信息
Analog processing has recently gained new traction as a result of developments in alternative computing paradigms to the von Neumann one. In particular, concerning opticalcomputing, neuromorphic inspired frameworks s...
详细信息
Diffractive neuralnetworks (D2NNs) have been transformative in many fields, motivating the development of various opticalcomputing components. However, these opticalcomputing components, achieving in diffractive op...
详细信息
In this work we discuss a diffractive opticalneural network approach for recognizing the mode of OAM waves and their superposition. Experimentally, diffractive neuralnetworks are fabricated through an imprinting tec...
详细信息
ISBN:
(数字)9798350370324
ISBN:
(纸本)9798350370331
In this work we discuss a diffractive opticalneural network approach for recognizing the mode of OAM waves and their superposition. Experimentally, diffractive neuralnetworks are fabricated through an imprinting technique with low loss parowax material. We also show that the diffractive networks can enable mathematical operations through the topological charges of the superposed OAM waves, being capable of displaying these results in a digit format across different operation planes. The approaches herein are general and could be used to arbitrarily manipulate multiple superposed OAM states, which can enable a myriad of potential applications in the next generation of terahertz communications systems.
Diffractive opticalneuralnetworks (DONNs) are emerging as high-throughput and energy-efficient hardware platforms to perform all-optical machine learning (ML) in machine vision systems. However, the current demonstr...
详细信息
Diffractive opticalneuralnetworks (DONNs) are emerging as high-throughput and energy-efficient hardware platforms to perform all-optical machine learning (ML) in machine vision systems. However, the current demonstrated applications of DONNs are largely image classification tasks, which undermine the prospect of developing and utilizing such hardware for other ML applications. Herein, the deployment of an all-optical reconfigurable DONNs system for scientific computing is demonstrated numerically and experimentally, including guiding two-dimensional quantum material synthesis, predicting the properties of two-dimensional quantum materials and small molecular cancer drugs, predicting the device response of nanopatterned integrated photonic power splitters, and the dynamic stabilization of an inverted pendulum with reinforcement learning. Despite a large variety of input data structures, a universal feature engineering approach is developed to convert categorical input features to images that can be processed in the DONNs system. The results open up new opportunities for employing DONNs systems for a broad range of ML applications. An all-optical reconfigurable diffractive opticalneural network system is deployed for performing scientific computing tasks, including guiding two-dimensional quantum material synthesis, predicting the properties of two-dimensional quantum materials and small molecular cancer drugs, predicting the device response of nanopatterned-integrated photonic power splitters, and the dynamic stabilization of an inverted pendulum with reinforcement *** (c) 2023 WILEY-VCH GmbH
In this paper we present a model to solve problems based on quantum machine learning approach using opticalneuralnetworks. Quantum computers and quantum-based machine learning approaches are gathering momentum in to...
详细信息
暂无评论