arXiv subjects

Xing Lin

Publications and source records attributed to Xing Lin.

34 records · Page 2Linked to original sources

arXiv2022

Optical multi-task learning using multi-wavelength diffractive deep neural networks

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.

cs.LG↗

arXiv2022

Optical Neural Ordinary Differential Equations

Increasing the layer number of on-chip photonic neural networks (PNNs) is essential to improve its model performance. However, the successively cascading of network hidden layers results in larger integrated photonic chip areas. To address this issue, we propose the optical neural ordinary differential equations (ON-ODE) architecture that parameterizes the continuous dynamics of hidden layers with optical ODE solvers. The ON-ODE comprises the PNNs followed by the photonic integrator and optical feedback loop, which can be configured to represent residual neural networks (ResNet) and recurrent neural networks with effectively reduced chip area occupancy. For the interference-based optoelectronic nonlinear hidden layer, the numerical experiments demonstrate that the single hidden layer ON-ODE can achieve approximately the same accuracy as the two-layer optical ResNet in image classification tasks. Besides, the ONODE improves the model classification accuracy for the diffraction-based all-optical linear hidden layer. The time-dependent dynamics property of ON-ODE is further applied for trajectory prediction with high accuracy.

cs.LG↗

arXiv2022

All-optical graph representation learning using integrated diffractive photonic computing units

Photonic neural networks perform brain-inspired computations using photons instead of electrons that can achieve substantially improved computing performance. However, existing architectures can only handle data with regular structures, e.g., images or videos, but fail to generalize to graph-structured data beyond Euclidean space, e.g., social networks or document co-citation networks. Here, we propose an all-optical graph representation learning architecture, termed diffractive graph neural network (DGNN), based on the integrated diffractive photonic computing units (DPUs) to address this limitation. Specifically, DGNN optically encodes node attributes into strip optical waveguides, which are transformed by DPUs and aggregated by on-chip optical couplers to extract their feature representations. Each DPU comprises successive passive layers of metalines to modulate the electromagnetic optical field via diffraction, where the metaline structures are learnable parameters shared across graph nodes. DGNN captures complex dependencies among the node neighborhoods and eliminates the nonlinear transition functions during the light-speed optical message passing over graph structures. We demonstrate the use of DGNN extracted features for node and graph-level classification tasks with benchmark databases and achieve superior performance. Our work opens up a new direction for designing application-specific integrated photonic circuits for high-efficiency processing of large-scale graph data structures using deep learning.

cs.ET↗

arXiv2022

Photonic unsupervised learning processor for secure and high-throughput optical fiber communication

Following the explosive growth of global data, there is an ever-increasing demand for high-throughput optical fiber communication (OFC) systems to perform massive data transmission and processing. Existing OFC methods mainly rely on electronic circuits for data processing, which severely limits the communication throughput. Though considered promising for the next-generation high-speed fiber communication, all-optical OFC remains unachievable due to serious challenges in effective optical computing, system modeling and configuring. Here we propose an end-to-end photonic encoder-decoder (PED) processor which maps the physical system of OFC into an optical generative neural network. By modeling the OFC transmission process as the variation in the constructed optical latent space, the PED learns noise-resistant coding schemes via unsupervised optimization. With multi-layer parametric diffractive neural networks, the PED establishes a large-scale and high-throughput optical computing framework that integrates the main OFC computations including coding, encryption and compression to the optical domain. The whole system improves the latency of computation in OFC systems by five orders of magnitude compared with the state-of-the-art device. On benchmarking datasets, the PED experimentally achieves up to 32% reduction in transmission error ratio (ER) than on-off keying (OOK), one of the mainstream methods with the lowest ER in general transmission. As we demonstrate on medical data, the PED increases the transmission throughput by two orders of magnitude than 8-level pulse amplitude modulation (PAM-8). We believe the proposed photonic encoder-decoder processor not only paves the way to the next-generation all-optical OFC systems, but also promotes a wide range of AI-based physical system designs.

physics.optics↗

arXiv2022

Double-Pulse Generation of Indistinguishable Single Photons with Optically Controlled Polarization

Single-photon sources play a key role in photonic quantum technologies. Semiconductor quantum dots can emit indistinguishable single photons under resonant excitation. However, the resonance fluorescence technique typically requires cross-polarization filtering which causes a loss of the unpolarized quantum dot emission by 50%. To solve this problem, we demonstrate a method to generate indistinguishable single photons with optically controlled polarization by two laser pulses off-resonant with neutral exciton states. This scheme is realized by exciting the quantum dot to the biexciton state and subsequently driving the quantum dot to an exciton eigenstate. Combining with magnetic field, we demonstrated the generation of photons with optically controlled polarization (polarization degree of 101(2)%), laser-neutral exciton detuning up to 0.81 meV, high single-photon purity (99.6(1)%) and indistinguishability (85(4)%). Laser pulses can be blocked using polarization and spectral filtering. Our work makes an important step towards indistinguishable single-photon sources with near-unity collection efficiency.

quant-ph↗

arXiv2022

A Class of Minimum Storage Cooperative Regenerating Codes with Low Access Property

In this paper, a new repair scheme for a modified construction of MDS codes is studied. The obtained repair scheme has optimal bandwidth for multiple failed nodes under the cooperative repair model. In addition, the repair scheme has relatively low access property, where the number of data accessed is less than two times the optimal value.

cs.IT↗

arXiv2021

Security Analysis and Improvement of Source Independent Quantum Random Number Generators with Imperfect Devices

A quantum random number generator (QRNG) as a genuine source of randomness is essential in many applications, such as number simulation and cryptography. Recently, a source-independent quantum random number generator (SI-QRNG), which can generate secure random numbers with untrusted sources, has been realized. However, the measurement loopholes of the trusted but imperfect devices used in SI-QRNGs have not yet been fully explored, which will cause security problems, especially in high-speed systems. Here, we point out and evaluate the security loopholes of practical imperfect measurement devices in SI-QRNGs. We also provide corresponding countermeasures to prevent these information leakages by recalculating the conditional minimum entropy and adding a monitor. Furthermore, by taking into account the finite-size effect,we show that the influence of the afterpulse can exceed that of the finite-size effect with the large number of sampled rounds. Our protocol is simple and effective, and it promotes the security of SI-QRNG in practice as well as the compatibility with high-speed measurement devices, thus paving the way for constructing ultrafast and security-certified commercial SI-QRNG systems.

quant-ph↗

arXiv2020

Observation of photon antibunching with a single conventional detector

The second-order photon correlation function is of great importance in quantum optics which is typically measured with the Hanbury Brown and Twiss interferometer which employs a pair of single-photon detectors and a dual-channel time acquisition module. Here we demonstrate a new method to measure and extract the second-order correlation function with a standard single-photon avalanche photodiode (dead-time = 22 ns) and a single-channel time acquisition module. This is realized by shifting the informative coincidence counts near the zero-time delay to a time window which is not obliterated by the dead-time and after-pulse of detection system. The new scheme is verified by measuring the second-order correlation from a single colloidal nanocrystal. Photon antibunching is unambiguously observed and agrees well with the result measured using the standard HBT setup. Our scheme simplifies the higher-order correlation technique and might be favored in cost-sensitive circumstances.

quant-ph↗

arXiv2020

Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit

Application-specific optical processors have been considered disruptive technologies for modern computing that can fundamentally accelerate the development of artificial intelligence (AI) by offering substantially improved computing performance. Recent advancements in optical neural network architectures for neural information processing have been applied to perform various machine learning tasks. However, the existing architectures have limited complexity and performance; and each of them requires its own dedicated design that cannot be reconfigured to switch between different neural network models for different applications after deployment. Here, we propose an optoelectronic reconfigurable computing paradigm by constructing a diffractive processing unit (DPU) that can efficiently support different neural networks and achieve a high model complexity with millions of neurons. It allocates almost all of its computational operations optically and achieves extremely high speed of data modulation and large-scale network parameter updating by dynamically programming optical modulators and photodetectors. We demonstrated the reconfiguration of the DPU to implement various diffractive feedforward and recurrent neural networks and developed a novel adaptive training approach to circumvent the system imperfections. We applied the trained networks for high-speed classifying of handwritten digit images and human action videos over benchmark datasets, and the experimental results revealed a comparable classification accuracy to the electronic computing approaches. Furthermore, our prototype system built with off-the-shelf optoelectronic components surpasses the performance of state-of-the-art graphics processing units (GPUs) by several times on computing speed and more than an order of magnitude on system energy efficiency.

eess.IV↗

arXiv2020

Deciphering exciton-generation processes in quantum-dot electroluminescence

Electroluminescence (EL) of colloidal nanocrystals promises a new generation of high-performance and solution-processable light-emitting diodes (LEDs). The operation of nanocrystal-based LEDs relies on the recombination of electrically-generated excitons. However, a fundamental question, i.e, how excitons are electrically generated in individual nanocrystals, remains unanswered. Here, we reveal a molecular mechanism of sequential electron-hole injection for the exciton generation in nanocrystal-based EL devices. To decipher the corresponding elementary processes, we develop electrically-pumped single-nanocrystal spectroscopy. While hole injection into neutral quantum dots (QDs) is generally-considered to be inefficient, we find that the intermediate negatively-charged state of QD triggers confinement-enhanced Coulomb interactions, which simultaneously accelerate hole injection and hinder excessive electron injection. In-situ/operando spectroscopy on state-of-the-art QD-LEDs demonstrate that exciton generation at the ensemble level is consistent with the charge-confinement-enabled sequential electron-hole injection mechanism revealed at the single-nanocrystal level. Our findings provide a universal mechanism for enhancing charge balance in nanocrystal-based EL devices.

physics.optics↗

arXiv2018

Response to Comment on "All-optical machine learning using diffractive deep neural networks"

In their Comment, Wei et al. (arXiv:1809.08360v1 [cs.LG]) claim that our original interpretation of Diffractive Deep Neural Networks (D2NN) represent a mischaracterization of the system due to linearity and passivity. In this Response, we detail how this mischaracterization claim is unwarranted and oblivious to several sections detailed in our original manuscript (Science, DOI: 10.1126/science.aat8084) that specifically introduced and discussed optical nonlinearities and reconfigurability of D2NNs, as part of our proposed framework to enhance its performance. To further refute the mischaracterization claim of Wei et al., we, once again, demonstrate the depth feature of optical D2NNs by showing that multiple diffractive layers operating collectively within a D2NN present additional degrees-of-freedom compared to a single diffractive layer to achieve better classification accuracy, as well as improved output signal contrast and diffraction efficiency as the number of diffractive layers increase, showing the deepness of a D2NN, and its inherent depth advantage for improved performance. In summary, the Comment by Wei et al. does not provide an amendment to the original teachings of our original manuscript, and all of our results, core conclusions and methodology of research reported in Science (DOI: 10.1126/science.aat8084) remain entirely valid.

cs.NE↗

arXiv2018

Optimal Resonant Beam Charging for Electronic Vehicles in Internet of Intelligent Vehicles

To enable electric vehicles (EVs) to access to the internet of intelligent vehicles (IoIV), charging EVs wirelessly anytime and anywhere becomes an urgent need. The resonant beam charging (RBC) technology can provide high-power and long-range wireless energy for EVs. However, the RBC system is unefficient. To improve the RBC power transmission efficiency, the adaptive resonant beam charging (ARBC) technology was introduced. In this paper, after analyzing the modular model of the ARBC system, we obtain the closed-form formula of the end-to-end power transmission efficiency. Then, we prove that the optimal power transmission efficiency uniquely exists. Moreover, we analyze the relationships among the optimal power transmission efficiency, the source power, the output power, and the beam transmission efficiency, which provide the guidelines for the optimal ARBC system design and implementation. Hence, perpetual energy can be supplied to EVs in IoIV virtually.

eess.SY↗

arXiv2018

All-Optical Machine Learning Using Diffractive Deep Neural Networks

We introduce an all-optical Diffractive Deep Neural Network (D2NN) architecture that can learn to implement various functions after deep learning-based design of passive diffractive layers that work collectively. We experimentally demonstrated the success of this framework by creating 3D-printed D2NNs that learned to implement handwritten digit classification and the function of an imaging lens at terahertz spectrum. With the existing plethora of 3D-printing and other lithographic fabrication methods as well as spatial-light-modulators, this all-optical deep learning framework can perform, at the speed of light, various complex functions that computer-based neural networks can implement, and will find applications in all-optical image analysis, feature detection and object classification, also enabling new camera designs and optical components that can learn to perform unique tasks using D2NNs.

cs.NE↗

arXiv2018

Extended depth-of-field in holographic image reconstruction using deep learning based auto-focusing and phase-recovery

Holography encodes the three dimensional (3D) information of a sample in the form of an intensity-only recording. However, to decode the original sample image from its hologram(s), auto-focusing and phase-recovery are needed, which are in general cumbersome and time-consuming to digitally perform. Here we demonstrate a convolutional neural network (CNN) based approach that simultaneously performs auto-focusing and phase-recovery to significantly extend the depth-of-field (DOF) in holographic image reconstruction. For this, a CNN is trained by using pairs of randomly de-focused back-propagated holograms and their corresponding in-focus phase-recovered images. After this training phase, the CNN takes a single back-propagated hologram of a 3D sample as input to rapidly achieve phase-recovery and reconstruct an in focus image of the sample over a significantly extended DOF. This deep learning based DOF extension method is non-iterative, and significantly improves the algorithm time-complexity of holographic image reconstruction from O(nm) to O(1), where n refers to the number of individual object points or particles within the sample volume, and m represents the focusing search space within which each object point or particle needs to be individually focused. These results highlight some of the unique opportunities created by data-enabled statistical image reconstruction methods powered by machine learning, and we believe that the presented approach can be broadly applicable to computationally extend the DOF of other imaging modalities.

cs.CV↗

arXiv2016

Influence of the Substrate Material on the Optical Properties of Tungsten Diselenide Monolayers

Monolayers of transition-metal dichalcogenides such as WSe2 have become increasingly attractive due to their potential in electrical and optical applications. Because the properties of these 2D systems are known to be affected by their surroundings, we report how the choice of the substrate material affects the optical properties of monolayer WSe2. To accomplish this study, pump-density-dependent micro-photoluminescence measurements are performed with time-integrating and time-resolving acquisition techniques. Spectral information and power-dependent mode intensities are compared at 290K and 10K for exfoliated WSe2 on SiO2/Si, sapphire (Al2O3), hBN/Si3N4/Si, and MgF2, indicating substrate-dependent appearance and strength of exciton, trion, and biexciton modes. Additionally, one CVD-grown WSe2 monolayer on sapphire is included in this study for direct comparison with its exfoliated counterpart. Time-resolved micro-photoluminescence shows how radiative decay times strongly differ for different substrate materials. Our data indicates exciton-exciton annihilation as a shortening mechanism at room temperature, and subtle trends in the decay rates in correlation to the dielectric environment at cryogenic temperatures. On the measureable time scales, trends are also related to the extent of the respective 2D-excitonic modes' appearance. This result highlights the importance of further detailed characterization of exciton features in 2D materials, particularly with respect to the choice of substrate.

cond-mat.mes-hall↗

arXiv2013

A Qualitative Representation and Similarity Measurement Method in Geographic Information Retrieval

The modern geographic information retrieval technology is based on quantitative models and methods. The semantic information in web documents and queries cannot be effectively represented, leading to information lost or misunderstanding so that the results are either unreliable or inconsistent. A new qualitative approach is thus proposed for supporting geographic information retrieval based on qualitative representation, semantic matching, and qualitative reasoning. A qualitative representation model and the corresponding similarity measurement method are defined. Information in documents and user queries are represented using propositional logic, which considers the thematic and geographic semantics synthetically. Thematic information is represented as thematic propositions on the base of domain ontology. Similarly, spatial information is represented as geo-spatial propositions with the support of geographic knowledge base. Then the similarity is divided into thematic similarity and spatial similarity. The former is calculated by the weighted distance of proposition keywords in the domain ontology, and the latter similarity is further divided into conceptual similarity and spatial similarity. Represented by propositions and information units, the similarity measurement can take evidence theory and fuzzy logic to combine all sub similarities to get the final similarity between documents and queries. This novel retrieval method is mainly used to retrieve the qualitative geographic information to support the semantic matching and results ranking. It does not deal with geometric computation and is consistent with human commonsense cognition, and thus can improve the efficiency of geographic information retrieval technology.

cs.IR↗