Deep learning has rapidly advanced amidst the proliferation of large models,leading to challenges in computational resources and power consumption.Optical neural networks(ONNs)offer a solution by shifting computation ...Deep learning has rapidly advanced amidst the proliferation of large models,leading to challenges in computational resources and power consumption.Optical neural networks(ONNs)offer a solution by shifting computation to optics,thereby leveraging the benefits of low power consumption,low latency,and high parallelism.The current training paradigm for ONNs primarily relies on backpropagation(BP).However,the reliance is incompatible with potential unknown processes within the system,which necessitates detailed knowledge and precise mathematical modeling of the optical process.In this paper,we present a pre-sensor multilayer ONN with nonlinear activation,utilizing a forward-forward algorithm to directly train both optical and digital parameters,which replaces the traditional backward pass with an additional forward pass.Our proposed nonlinear optical system demonstrates significant improvements in image classification accuracy,achieving a maximum enhancement of 9.0%.It also validates the efficacy of training parameters in the presence of unknown nonlinear components in the optical system.The proposed training method addresses the limitations of BP,paving the way for applications with a broader range of physical transformations in ONNs.展开更多
Based on Presnel-Kirchhoff diffraction theory, we set up a diffraction model of nonlinear optical media to Gaussian beam, which can interpret the Z-scan phenomenon from a new way. This theory is not only well consiste...Based on Presnel-Kirchhoff diffraction theory, we set up a diffraction model of nonlinear optical media to Gaussian beam, which can interpret the Z-scan phenomenon from a new way. This theory is not only well consistent with the conventional Z-scan theory in the case of small nonlinear phase shift, but also can fit for the lager nonlinear phase shift. Numeric computations indicate the shape of the Z-scan curve is greatly affected by the value of the nonlinear phase shift. The symmetric dispersion-like Z-scan curve is only valid for small nonlinear phase shift (|Δφ0| < π), but with increasing the nonlinear phase shift, the valley of the transmittance is severely suppressed and the peak is greatly enhanced. Further calculations show some new interesting results.展开更多
基金National Key Research and Development Program of China(2024YFE0203600)National Natural Science Foundation of China(62135009)Beijing Municipal Science&Technology Commission,Administrative Commission of Zhongguancun Science Park(Z221100005322010)。
文摘Deep learning has rapidly advanced amidst the proliferation of large models,leading to challenges in computational resources and power consumption.Optical neural networks(ONNs)offer a solution by shifting computation to optics,thereby leveraging the benefits of low power consumption,low latency,and high parallelism.The current training paradigm for ONNs primarily relies on backpropagation(BP).However,the reliance is incompatible with potential unknown processes within the system,which necessitates detailed knowledge and precise mathematical modeling of the optical process.In this paper,we present a pre-sensor multilayer ONN with nonlinear activation,utilizing a forward-forward algorithm to directly train both optical and digital parameters,which replaces the traditional backward pass with an additional forward pass.Our proposed nonlinear optical system demonstrates significant improvements in image classification accuracy,achieving a maximum enhancement of 9.0%.It also validates the efficacy of training parameters in the presence of unknown nonlinear components in the optical system.The proposed training method addresses the limitations of BP,paving the way for applications with a broader range of physical transformations in ONNs.
基金This work was supported by the National Natural Science Foundation of China under Grant No. 60007009 the President Foundation of Chinese Academy of Sciences under Grant No. 40007059.
文摘Based on Presnel-Kirchhoff diffraction theory, we set up a diffraction model of nonlinear optical media to Gaussian beam, which can interpret the Z-scan phenomenon from a new way. This theory is not only well consistent with the conventional Z-scan theory in the case of small nonlinear phase shift, but also can fit for the lager nonlinear phase shift. Numeric computations indicate the shape of the Z-scan curve is greatly affected by the value of the nonlinear phase shift. The symmetric dispersion-like Z-scan curve is only valid for small nonlinear phase shift (|Δφ0| < π), but with increasing the nonlinear phase shift, the valley of the transmittance is severely suppressed and the peak is greatly enhanced. Further calculations show some new interesting results.
