Photonic neural networks are brain-inspired information processing technology using photons instead of electrons to perform artificial intelligence (AI) tasks. However, existing architectures are designed for a single task but fail to multiplex different tasks in parallel within a single monolithic system due to the task competition that deteriorates the model performance. This paper proposes a novel optical multi-task learning system by designing multi-wavelength diffractive deep neural networks (D2NNs) with the joint optimization method. By encoding multi-task inputs into multi-wavelength channels, the system can increase the computing throughput and significantly alle-viate the competition to perform multiple tasks in parallel with high accuracy. We design the two-task and four-task D2NNs with two and four spectral channels, respectively, for classifying different inputs from MNIST, FMNIST, KMNIST, and EMNIST databases. The numerical evaluations demonstrate that, under the same network size, mul-ti-wavelength D2NNs achieve significantly higher classification accuracies for multi-task learning than single-wavelength D2NNs. Furthermore, by increasing the network size, the multi-wavelength D2NNs for simultaneously performing multiple tasks achieve comparable classification accuracies with respect to the individual training of multiple single-wavelength D2NNs to perform tasks separately. Our work paves the way for developing the wave-length-division multiplexing technology to achieve high-throughput neuromorphic photonic computing and more general AI systems to perform multiple tasks in parallel.

随机且未知的散射介质背后的对象的分类为计算成像和机器视野字段的具有挑战性的任务。最新的基于深度学习的方法证明了使用图像传感器收集的扩散器延伸模式对对象进行分类。这些方法需要使用在数字计算机上运行的深神经网络进行相对大规模的计算。在这里，我们提出了一个全光处理器，使用单个像素检测到的宽带照明通过未知的随机相扩散器直接对未知对象进行分类。使用深度学习进行了优化的一组传播衍射层，形成了一个物理网络，该物理网络全面地绘制了随机扩散器后面输入对象的空间信息，以进入通过单个像素在输出平面上检测到的输出光的功率谱，衍射网络。我们在数值上使用宽带辐射通过随机新扩散器对未知手写数字进行分类，在训练阶段从未使用过，并实现了88.53％的盲目测试准确性。这种通过随机扩散器的单像素全光对象分类系统基于被动衍射层，该层可以通过简单地缩放与波长范围的衍射范围来缩放衍射特征，从而在电磁光谱的任何部分中运行，并且可以在电磁光谱的任何部分工作。这些结果在例如生物医学成像，安全性，机器人技术和自动驾驶中具有各种潜在的应用。

Diffractive optical networks provide rich opportunities for visual computing tasks since the spatial information of a scene can be directly accessed by a diffractive processor without requiring any digital pre-processing steps. Here we present data class-specific transformations all-optically performed between the input and output fields-of-view (FOVs) of a diffractive network. The visual information of the objects is encoded into the amplitude (A), phase (P), or intensity (I) of the optical field at the input, which is all-optically processed by a data class-specific diffractive network. At the output, an image sensor-array directly measures the transformed patterns, all-optically encrypted using the transformation matrices pre-assigned to different data classes, i.e., a separate matrix for each data class. The original input images can be recovered by applying the correct decryption key (the inverse transformation) corresponding to the matching data class, while applying any other key will lead to loss of information. The class-specificity of these all-optical diffractive transformations creates opportunities where different keys can be distributed to different users; each user can only decode the acquired images of only one data class, serving multiple users in an all-optically encrypted manner. We numerically demonstrated all-optical class-specific transformations covering A-->A, I-->I, and P-->I transformations using various image datasets. We also experimentally validated the feasibility of this framework by fabricating a class-specific I-->I transformation diffractive network using two-photon polymerization and successfully tested it at 1550 nm wavelength. Data class-specific all-optical transformations provide a fast and energy-efficient method for image and data encryption, enhancing data security and privacy.

衍射深神经网络（D2NNS）定义了一个由空间工程的被动表面组成的全光计算框架，该框架通过调节传播光的幅度和/或相位来共同处理光学输入信息。衍射光学网络通过薄衍射量以光的速度来完成其计算任务，而无需任何外部计算能力，同时利用了光学的巨大并行性。证明了衍射网络以实现对象的全光分类并执行通用线性变换。在这里，我们首次证明了使用衍射网络的“延时”图像分类方案，通过使用输入对象的横向运动和/或衍射网络，可以显着提高其在复杂输入对象上的分类准确性和概括性性能。 ，相对于彼此。在不同的上下文中，通常将对象和/或相机的相对运动用于图像超分辨率应用程序；受其成功的启发，我们设计了一个延时衍射网络，以受益于由受控或随机横向移动创建的互补信息内容。我们从数值探索了延时衍射网络的设计空间和性能限制，从CIFAR-10数据集的对象进行光学分类中揭示了62.03％的盲测精度。这构成了迄今使用CIFAR-10数据集上的单个衍射网络达到的最高推理精度。延时衍射网络将对使用全光处理器的输入信号的时空分析广泛有用。

增加片上光子神经网络（PNN）的层数对于改善其模型性能至关重要。但是，网络隐藏层的连续级联导致更大的集成光子芯片区域。为了解决此问题，我们提出了光学神经常规微分方程（ON-ON-ON-OD-ON-OD-ON-OD-ON-OD-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ODINE），该架构用光ODE求解器参数化了隐藏层的连续动力学。 On-Ode包括PNN，然后是光子积分器和光反馈回路，可以配置为代表残留的神经网络（RESNET）和复发性神经网络，并有效地降低了芯片面积占用率。对于基于干扰的光电非线性隐藏层，数值实验表明，单个隐藏层ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ON-ONE表示与图像分类任务中的两层光学重新系统大致相同。此外，Onode提高了基于衍射的全光线性隐藏层的模型分类精度。 On-Eod的时间依赖性动力学属性进一步应用于高精度的轨迹预测。

随着深度神经网络（DNN）的发展以解决日益复杂的问题，它们正受到现有数字处理器的延迟和功耗的限制。为了提高速度和能源效率，已经提出了专门的模拟光学和电子硬件，但是可扩展性有限（输入矢量长度$ k $的数百个元素）。在这里，我们提出了一个可扩展的，单层模拟光学处理器，该光学处理器使用自由空间光学器件可重新配置输入向量和集成的光电，用于静态，可更新的加权和非线性 - 具有$ k \ \ 1,000 $和大约1,000美元和超过。我们通过实验测试MNIST手写数字数据集的分类精度，在没有数据预处理或在硬件上进行数据重新处理的情况下达到94.7％（地面真相96.3％）。我们还确定吞吐量（$ \ sim $ 0.9 examac/s）的基本上限，由最大光带宽设置，然后大大增加误差。我们在兼容CMOS兼容系统中宽光谱和空间带宽的组合可以实现下一代DNN的高效计算。

由于深度学习在许多人工智能应用中显示了革命性的性能，其升级的计算需求需要用于巨大并行性的硬件加速器和改进的吞吐量。光学神经网络（ONN）是下一代神经关键组成的有希望的候选者，由于其高并行，低延迟和低能量消耗。在这里，我们设计了一个硬件高效的光子子空间神经网络（PSNN）架构，其针对具有比具有可比任务性能的前一个ONN架构的光学元件使用，区域成本和能量消耗。此外，提供了一种硬件感知培训框架，以最小化所需的设备编程精度，减少芯片区域，并提高噪声鲁棒性。我们在实验上展示了我们的PSNN在蝴蝶式可编程硅光子集成电路上，并在实用的图像识别任务中显示其实用性。

Machine learning methods have revolutionized the discovery process of new molecules and materials. However, the intensive training process of neural networks for molecules with ever-increasing complexity has resulted in exponential growth in computation cost, leading to long simulation time and high energy consumption. Photonic chip technology offers an alternative platform for implementing neural networks with faster data processing and lower energy usage compared to digital computers. Photonics technology is naturally capable of implementing complex-valued neural networks at no additional hardware cost. Here, we demonstrate the capability of photonic neural networks for predicting the quantum mechanical properties of molecules. To the best of our knowledge, this work is the first to harness photonic technology for machine learning applications in computational chemistry and molecular sciences, such as drug discovery and materials design. We further show that multiple properties can be learned simultaneously in a photonic chip via a multi-task regression learning algorithm, which is also the first of its kind as well, as most previous works focus on implementing a network in the classification task.

置换矩阵构成了一个重要的计算构建块，这些构建块在各个领域中经常使用，例如通信，信息安全和数据处理。具有相对较大数量的基于功率，快速和紧凑型平台的输入输出互连的置换运算符的光学实现是非常可取的。在这里，我们提出了通过深度学习设计的衍射光学网络，以全面执行置换操作，可以使用被动的传播层在输入和视场之间扩展到数十万个互连，这些互连是在波长规模上单独构造的。 。我们的发现表明，衍射光网络在近似给定置换操作中的容量与系统中衍射层和可训练的传输元件的数量成正比。这种更深的衍射网络设计可以在系统的物理对齐和输出衍射效率方面构成实际挑战。我们通过设计不对对准的衍射设计来解决这些挑战，这些设计可以全面执行任意选择的置换操作，并首次在实验中证明了在频谱的THZ部分运行的衍射排列网络。衍射排列网络可能会在例如安全性，图像加密和数据处理以及电信中找到各种应用程序；尤其是在无线通信中的载波频率接近THZ波段的情况下，提出的衍射置换网络可以潜在地充当无线网络中的通道路由和互连面板。

Ever since the first microscope by Zacharias Janssen in the late 16th century, scientists have been inventing new types of microscopes for various tasks. Inventing a novel architecture demands years, if not decades, worth of scientific experience and creativity. In this work, we introduce Differentiable Microscopy ($\partial\mu$), a deep learning-based design paradigm, to aid scientists design new interpretable microscope architectures. Differentiable microscopy first models a common physics-based optical system however with trainable optical elements at key locations on the optical path. Using pre-acquired data, we then train the model end-to-end for a task of interest. The learnt design proposal can then be simplified by interpreting the learnt optical elements. As a first demonstration, based on the optical 4-$f$ system, we present an all-optical quantitative phase microscope (QPM) design that requires no computational post-reconstruction. A follow-up literature survey suggested that the learnt architecture is similar to the generalized phase contrast method developed two decades ago. Our extensive experiments on multiple datasets that include biological samples show that our learnt all-optical QPM designs consistently outperform existing methods. We experimentally verify the functionality of the optical 4-$f$ system based QPM design using a spatial light modulator. Furthermore, we also demonstrate that similar results can be achieved by an uninterpretable learning based method, namely diffractive deep neural networks (D2NN). The proposed differentiable microscopy framework supplements the creative process of designing new optical systems and would perhaps lead to unconventional but better optical designs.

计算光学成像（COI）系统利用其设置中的光学编码元素（CE）在单个或多个快照中编码高维场景，并使用计算算法对其进行解码。 COI系统的性能很大程度上取决于其主要组件的设计：CE模式和用于执行给定任务的计算方法。常规方法依赖于随机模式或分析设计来设置CE的分布。但是，深神经网络（DNNS）的可用数据和算法功能已在CE数据驱动的设计中开辟了新的地平线，该设计共同考虑了光学编码器和计算解码器。具体而言，通过通过完全可区分的图像形成模型对COI测量进行建模，该模型考虑了基于物理的光及其与CES的相互作用，可以在端到端优化定义CE和计算解码器的参数和计算解码器（e2e）方式。此外，通过在同一框架中仅优化CE，可以从纯光学器件中执行推理任务。这项工作调查了CE数据驱动设计的最新进展，并提供了有关如何参数化不同光学元素以将其包括在E2E框架中的指南。由于E2E框架可以通过更改损耗功能和DNN来处理不同的推理应用程序，因此我们提出低级任务，例如光谱成像重建或高级任务，例如使用基于任务的光学光学体系结构来增强隐私的姿势估计，以维护姿势估算。最后，我们说明了使用全镜DNN以光速执行的分类和3D对象识别应用程序。

The ever-growing deep learning technologies are making revolutionary changes for modern life. However, conventional computing architectures are designed to process sequential and digital programs, being extremely burdened with performing massive parallel and adaptive deep learning applications. Photonic integrated circuits provide an efficient approach to mitigate bandwidth limitations and power-wall brought by its electronic counterparts, showing great potential in ultrafast and energy-free high-performance computing. Here, we propose an optical computing architecture enabled by on-chip diffraction to implement convolutional acceleration, termed optical convolution unit (OCU). We demonstrate that any real-valued convolution kernels can be exploited by OCU with a prominent computational throughput boosting via the concept of structral re-parameterization. With OCU as the fundamental unit, we build an optical convolutional neural network (oCNN) to implement two popular deep learning tasks: classification and regression. For classification, Fashion-MNIST and CIFAR-4 datasets are tested with accuracy of 91.63% and 86.25%, respectively. For regression, we build an optical denoising convolutional neural network (oDnCNN) to handle Gaussian noise in gray scale images with noise level {\sigma} = 10, 15, 20, resulting clean images with average PSNR of 31.70dB, 29.39dB and 27.72dB, respectively. The proposed OCU presents remarkable performance of low energy consumption and high information density due to its fully passive nature and compact footprint, providing a highly parallel while lightweight solution for future computing architecture to handle high dimensional tensors in deep learning.

近年来，人工智能（AI）的领域已经见证了巨大的增长，然而，持续发展的一些最紧迫的挑战是电子计算机架构所面临的基本带宽，能效和速度限制。利用用于执行神经网络推理操作的光子处理器越来越感兴趣，但是这些网络目前使用标准数字电子培训。这里，我们提出了由CMOS兼容的硅光子架构实现的神经网络的片上训练，以利用大规模平行，高效和快速数据操作的电位。我们的方案采用直接反馈对准训练算法，它使用错误反馈而不是错误反向化而培训神经网络，并且可以在每秒乘以数万亿乘以量的速度运行，同时每次MAC操作消耗小于一个微微约会。光子架构利用并行化矩阵 - 向量乘法利用微址谐振器阵列，用于沿着单个波导总线处理多通道模拟信号，以便原位计算每个神经网络层的梯度向量，这是在后向通过期间执行的最昂贵的操作。 。我们还通过片上MAC操作结果实验地示意使用MNIST数据集进行培训深度神经网络。我们的高效，超快速神经网络训练的新方法展示了光子学作为执行AI应用的有希望的平台。

由于其潜在的优势，如可扩展性，低延迟和功率效率，光学计算在过去几十年中已经看到了快速进步。潜在的全光处理器的核心单元将是NAND门，其可以级联以执行任意逻辑操作。在这里，我们使用衍射神经网络呈现可级可级联的全光NAND门的设计和分析。我们使用两个空间分离的孔的相对光功率在衍射NAND门的输入和输出平面上进行了编码的逻辑值。基于该架构，我们使用光的衍射来了数值优化了由4个无源层组成的衍射神经网络的设计，并通过光的衍射来实现这些衍射NAND操作，通过连续地馈送输出来级联这些衍射NAND门来执行复杂的逻辑功能一个衍射NAND门进入另一个。我们通过使用相同的衍射设计来显示我们的衍射NAND门的级联性，以及全光学执行和或操作以及半加法器。由空间工程化无源衍射层组成的可级可抵级光学NAND门可以用作各种光学计算平台的核心组件。

与常规深层神经网络（DNN）相比，衍射光学神经网络（DONNS）在功率效率，并行性和计算速度方面具有显着优势，因此引起了很多关注，这些神经网络（DNN）在数字平台上实现时具有内在的限制。但是，反相反的算法训练的物理模型参数上具有离散值的现实世界光学设备是一个非平凡的任务，因为现有的光学设备具有非统一的离散级别和非单调属性。这项工作提出了一个新颖的设备对系统硬件软件代码框架，该框架可以对Donns W.R.T的有效物理意识培训进行跨层的任意实验测量的光学设备。具体而言，使用Gumbel-SoftMax来启用从现实世界设备参数的可区分映射到Donns的正向函数，在Donn中，Donn中的物理参数可以通过简单地最小化ML任务的损耗函数来训练。结果表明，我们提出的框架比传统的基于基于量化的方法具有显着优势，尤其是使用低精确的光学设备。最后，在低精度设置中，通过物理实验光学系统对所提出的算法进行了充分的验证。

A unidirectional imager would only permit image formation along one direction, from an input field-of-view (FOV) A to an output FOV B, and in the reverse path, the image formation would be blocked. Here, we report the first demonstration of unidirectional imagers, presenting polarization-insensitive and broadband unidirectional imaging based on successive diffractive layers that are linear and isotropic. These diffractive layers are optimized using deep learning and consist of hundreds of thousands of diffractive phase features, which collectively modulate the incoming fields and project an intensity image of the input onto an output FOV, while blocking the image formation in the reverse direction. After their deep learning-based training, the resulting diffractive layers are fabricated to form a unidirectional imager. As a reciprocal device, the diffractive unidirectional imager has asymmetric mode processing capabilities in the forward and backward directions, where the optical modes from B to A are selectively guided/scattered to miss the output FOV, whereas for the forward direction such modal losses are minimized, yielding an ideal imaging system between the input and output FOVs. Although trained using monochromatic illumination, the diffractive unidirectional imager maintains its functionality over a large spectral band and works under broadband illumination. We experimentally validated this unidirectional imager using terahertz radiation, very well matching our numerical results. Using the same deep learning-based design strategy, we also created a wavelength-selective unidirectional imager, where two unidirectional imaging operations, in reverse directions, are multiplexed through different illumination wavelengths. Diffractive unidirectional imaging using structured materials will have numerous applications in e.g., security, defense, telecommunications and privacy protection.

具有最小延迟的人工神经网络的决策对于诸如导航，跟踪和实时机器动作系统之类的许多应用来说是至关重要的。这要求机器学习硬件以高吞吐量处理多维数据。不幸的是，处理卷积操作是数据分类任务的主要计算工具，遵循有挑战性的运行时间复杂性缩放法。然而，在傅立叶光学显示器 - 光处理器中同心地实现卷积定理，使得不迭代的O（1）运行时复杂度以超过1,000×1,000大矩阵的数据输入。在此方法之后，这里我们展示了具有傅里叶卷积神经网络（FCNN）加速器的数据流多核图像批处理。我们将大规模矩阵的图像批量处理显示为傅立叶域中的数字光处理模块执行的被动的2000万点产品乘法。另外，我们通过利用多种时空衍射令并进一步并行化该光学FCNN系统，从而实现了最先进的FCNN加速器的98倍的产量改进。综合讨论与系统能力边缘工作相关的实际挑战突出了傅立叶域和决议缩放法律的串扰问题。通过利用展示技术中的大规模平行性加速卷积带来了基于VAN Neuman的机器学习加速度。

The ultimate goal of artificial intelligence is to mimic the human brain to perform decision-making and control directly from high-dimensional sensory input. All-optical diffractive neural networks provide a promising solution for realizing artificial intelligence with high-speed and low-power consumption. To date, most of the reported diffractive neural networks focus on single or multiple tasks that do not involve interaction with the environment, such as object recognition and image classification, while the networks that can perform decision-making and control, to our knowledge, have not been developed yet. Here, we propose to use deep reinforcement learning to realize diffractive neural networks that enable imitating the human-level capability of decision-making and control. Such networks allow for finding optimal control policies through interaction with the environment and can be readily realized with the dielectric metasurfaces. The superior performances of these networks are verified by engaging three types of classic games, Tic-Tac-Toe, Super Mario Bros., and Car Racing, and achieving the same or even higher levels comparable to human players. Our work represents a solid step of advancement in diffractive neural networks, which promises a fundamental shift from the target-driven control of a pre-designed state for simple recognition or classification tasks to the high-level sensory capability of artificial intelligence. It may find exciting applications in autonomous driving, intelligent robots, and intelligent manufacturing.

Multispectral imaging has been used for numerous applications in e.g., environmental monitoring, aerospace, defense, and biomedicine. Here, we present a diffractive optical network-based multispectral imaging system trained using deep learning to create a virtual spectral filter array at the output image field-of-view. This diffractive multispectral imager performs spatially-coherent imaging over a large spectrum, and at the same time, routes a pre-determined set of spectral channels onto an array of pixels at the output plane, converting a monochrome focal plane array or image sensor into a multispectral imaging device without any spectral filters or image recovery algorithms. Furthermore, the spectral responsivity of this diffractive multispectral imager is not sensitive to input polarization states. Through numerical simulations, we present different diffractive network designs that achieve snapshot multispectral imaging with 4, 9 and 16 unique spectral bands within the visible spectrum, based on passive spatially-structured diffractive surfaces, with a compact design that axially spans ~72 times the mean wavelength of the spectral band of interest. Moreover, we experimentally demonstrate a diffractive multispectral imager based on a 3D-printed diffractive network that creates at its output image plane a spatially-repeating virtual spectral filter array with 2x2=4 unique bands at terahertz spectrum. Due to their compact form factor and computation-free, power-efficient and polarization-insensitive forward operation, diffractive multispectral imagers can be transformative for various imaging and sensing applications and be used at different parts of the electromagnetic spectrum where high-density and wide-area multispectral pixel arrays are not widely available.

使用单像素检测，联合优化编码和解码的端到端神经网络可以实现高精度成像和高电平语义传感。然而，对于不同的采样率，大规模网络需要重新培训，这是呈现的呈现和计算消耗。在这封信中，我们报告了一种加权优化技术，用于动态速率自适应单像素成像和感应，只需要培训网络一次可用于任何采样率的时间一次。具体地，我们在编码过程中引入一种新的加权方案，以表征不同的模式的调制效率。虽然网络以高采样速率训练，但是迭代地更新调制模式和相应的权重，这在融合时产生最佳排名编码串。在实验实施方案中，采用最高重量的最佳模式系列用于光调制，从而实现高效的成像和感测。报告的策略节省了现有动态单像素网络所需另一种低速速率网络的额外培训，这进一步加倍训练效率。验证了Mnist DataSet上的实验，通过采样率为1的网络培训，平均成像PSNR为0.1采样率达到23.50 dB，并且图像的图像分类精度达到高达95.00 \％，以0.03的采样率达到95.00 \％ 97.91 \％以0.1的采样率。