This study introduces an optical neural network(ONN)-based autoencoder for efficient image processing,utilizing specialized optical matrix-vector multipliers for both encoding and decoding tasks.To address the challen...This study introduces an optical neural network(ONN)-based autoencoder for efficient image processing,utilizing specialized optical matrix-vector multipliers for both encoding and decoding tasks.To address the challenges in efficient decoding,we propose a method that optimizes output processing through scalar multiplications,enhancing performance in generating higher-dimensional outputs.By employing on-system iterative tuning,we mitigate hardware imperfections and noise,progressively improving image reconstruction accuracy to near-digital quality.Furthermore,our approach supports noise reduction and optical image generation,enabling models such as denoising autoencoders,variational autoencoders,and generative adversarial networks.Our results demonstrate that ONN-based systems have the potential to surpass the energy efficiency of traditional electronic systems,enabling real-time,low-power image processing in applications such as medical imaging,autonomous vehicles,and edge computing.展开更多
Photonics is promising to handle extensive vector multiplications in artificial intelligence(AI)techniques due to natural bosonic parallelism and high-speed information transmission.However,the dimensionality of curre...Photonics is promising to handle extensive vector multiplications in artificial intelligence(AI)techniques due to natural bosonic parallelism and high-speed information transmission.However,the dimensionality of current photonic linear operation is limited and tough to improve due to the complex beam interaction for implementing optical matrix operation and digital-analog conversions.Here,we propose a programmable and reconfigurable photonic linear vector machine with extreme scalability formed by a series of emitter-detector pairs as the independent basic computing units.The elemental values of two high-dimensional vectors are prepared on emitter-detector pairs by bit encoding and analog detecting method without requiring large-scale analog-to-digital converter or digital-to-analog converter arrays.Since there is no interaction among light beams inside,extreme scalability could be achieved by simply multiplicating the independent emitter-detector pair.The proposed architecture is inspired by the traditional Chinese Suanpan or abacus,and thus is denoted as photonic SUANPAN.Experimentally,the computing fidelities for vector inner products could achieve>98%in our implementation with an 8×8 vertical cavity surface emission laser(VCSEL)array and an 8×8 MoTe_(2)two-dimensional material photodetector array.Furthermore,such implementation is applied on two typical AI tasks as 1024-dimensional optimization problem is successfully solved and competitive classification accuracy of 88%is achieved for handwritten digit dataset.We believe that the photonic SUANPAN could serve as a fundamental linear vector machine and enhance various future AI applications.展开更多
In recent years,the explosive development of artificial intelligence implementing by artificial neural networks(ANNs)creates inconceivable demands for computing hardware.However,conventional computing hardware based o...In recent years,the explosive development of artificial intelligence implementing by artificial neural networks(ANNs)creates inconceivable demands for computing hardware.However,conventional computing hardware based on electronic transistor and von Neumann architecture cannot satisfy such an inconceivable demand due to the unsustainability of Moore’s Law and the failure of Dennard’s scaling rules.Fortunately,analog optical computing offers an alternative way to release unprecedented computational capability to accelerate varies computing drained tasks.In this article,the challenges of the modern computing technologies and potential solutions are briefly explained in Chapter 1.In Chapter 2,the latest research progresses of analog optical computing are separated into three directions:vector/matrix manipulation,reservoir computing and photonic Ising machine.Each direction has been explicitly summarized and discussed.The last chapter explains the prospects and the new challenges of analog optical computing.展开更多
Conventional electronic processors,which are the mainstream and almost invincible hardware for computation,are approaching their limits in both computational power and energy efficiency,especially in large-scale matri...Conventional electronic processors,which are the mainstream and almost invincible hardware for computation,are approaching their limits in both computational power and energy efficiency,especially in large-scale matrix computation.By combining electronic,photonic,and optoelectronic devices and circuits together,silicon-based optoelectronic matrix computation has been demonstrating great capabilities and feasibilities.Matrix computation is one of the few general-purpose computations that have the potential to exceed the computation performance of digital logic circuits in energy efficiency,computational power,and latency.Moreover,electronic processors also suffer from the tremendous energy consumption of the digital transceiver circuits during high-capacity data interconnections.We review the recent progress in photonic matrix computation,including matrix-vector multiplication,convolution,and multiply–accumulate operations in artificial neural networks,quantum information processing,combinatorial optimization,and compressed sensing,with particular attention paid to energy consumption.We also summarize the advantages of siliconbased optoelectronic matrix computation in data interconnections and photonic-electronic integration over conventional optical computing processors.Looking toward the future of silicon-based optoelectronic matrix computations,we believe that silicon-based optoelectronics is a promising and comprehensive platform for disruptively improving general-purpose matrix computation performance in the post-Moore’s law era.展开更多
基金supported by the National Research Foundation(NRF)of Korea funded by the Ministry of Science,ICT and Future Planning(MSIP)of Korea(Nos.RS-2021-NR060087 and RS-2024-00353762)。
文摘This study introduces an optical neural network(ONN)-based autoencoder for efficient image processing,utilizing specialized optical matrix-vector multipliers for both encoding and decoding tasks.To address the challenges in efficient decoding,we propose a method that optimizes output processing through scalar multiplications,enhancing performance in generating higher-dimensional outputs.By employing on-system iterative tuning,we mitigate hardware imperfections and noise,progressively improving image reconstruction accuracy to near-digital quality.Furthermore,our approach supports noise reduction and optical image generation,enabling models such as denoising autoencoders,variational autoencoders,and generative adversarial networks.Our results demonstrate that ONN-based systems have the potential to surpass the energy efficiency of traditional electronic systems,enabling real-time,low-power image processing in applications such as medical imaging,autonomous vehicles,and edge computing.
基金Funding from the National Key Research and Development Program of China(2023YFB2806703)the National Natural Science Foundation of China(Grant Nos.U22A6004,92365210,and 62175124)is greatly acknowledgedsupported by Beijing National Research Center for Information Science and Technology(BNRist),Frontier Science Center for Quantum Information,Beijing Academy of Quantum Information Science,and Tsinghua University Initiative Scientific Research Program.
文摘Photonics is promising to handle extensive vector multiplications in artificial intelligence(AI)techniques due to natural bosonic parallelism and high-speed information transmission.However,the dimensionality of current photonic linear operation is limited and tough to improve due to the complex beam interaction for implementing optical matrix operation and digital-analog conversions.Here,we propose a programmable and reconfigurable photonic linear vector machine with extreme scalability formed by a series of emitter-detector pairs as the independent basic computing units.The elemental values of two high-dimensional vectors are prepared on emitter-detector pairs by bit encoding and analog detecting method without requiring large-scale analog-to-digital converter or digital-to-analog converter arrays.Since there is no interaction among light beams inside,extreme scalability could be achieved by simply multiplicating the independent emitter-detector pair.The proposed architecture is inspired by the traditional Chinese Suanpan or abacus,and thus is denoted as photonic SUANPAN.Experimentally,the computing fidelities for vector inner products could achieve>98%in our implementation with an 8×8 vertical cavity surface emission laser(VCSEL)array and an 8×8 MoTe_(2)two-dimensional material photodetector array.Furthermore,such implementation is applied on two typical AI tasks as 1024-dimensional optimization problem is successfully solved and competitive classification accuracy of 88%is achieved for handwritten digit dataset.We believe that the photonic SUANPAN could serve as a fundamental linear vector machine and enhance various future AI applications.
文摘In recent years,the explosive development of artificial intelligence implementing by artificial neural networks(ANNs)creates inconceivable demands for computing hardware.However,conventional computing hardware based on electronic transistor and von Neumann architecture cannot satisfy such an inconceivable demand due to the unsustainability of Moore’s Law and the failure of Dennard’s scaling rules.Fortunately,analog optical computing offers an alternative way to release unprecedented computational capability to accelerate varies computing drained tasks.In this article,the challenges of the modern computing technologies and potential solutions are briefly explained in Chapter 1.In Chapter 2,the latest research progresses of analog optical computing are separated into three directions:vector/matrix manipulation,reservoir computing and photonic Ising machine.Each direction has been explicitly summarized and discussed.The last chapter explains the prospects and the new challenges of analog optical computing.
基金supported by the National Natural Science Foundation of China(62035001 and 61775005)。
文摘Conventional electronic processors,which are the mainstream and almost invincible hardware for computation,are approaching their limits in both computational power and energy efficiency,especially in large-scale matrix computation.By combining electronic,photonic,and optoelectronic devices and circuits together,silicon-based optoelectronic matrix computation has been demonstrating great capabilities and feasibilities.Matrix computation is one of the few general-purpose computations that have the potential to exceed the computation performance of digital logic circuits in energy efficiency,computational power,and latency.Moreover,electronic processors also suffer from the tremendous energy consumption of the digital transceiver circuits during high-capacity data interconnections.We review the recent progress in photonic matrix computation,including matrix-vector multiplication,convolution,and multiply–accumulate operations in artificial neural networks,quantum information processing,combinatorial optimization,and compressed sensing,with particular attention paid to energy consumption.We also summarize the advantages of siliconbased optoelectronic matrix computation in data interconnections and photonic-electronic integration over conventional optical computing processors.Looking toward the future of silicon-based optoelectronic matrix computations,we believe that silicon-based optoelectronics is a promising and comprehensive platform for disruptively improving general-purpose matrix computation performance in the post-Moore’s law era.
