基于System Generator的卷积加速结构设计与实现

为解决卷积神经网络中卷积运算耗时长、运算复杂的问题,针对卷积运算的并行性特征,提出一种基于分块的流水线加速方法,并基于该方法在System Generator上进行了电路设计。通过在FPGA(Field-programmable Gate Array)上进行实验验证,该设计模型能正确输出卷积运算结果;在结构和输入数据相同的情况下,该设计模型在计算速度上相比于普通CPU最高可加速258倍,相比于服务器级CPU提高了近40倍,具有良好的加速效果。

Computational Experiments for Complex Social Systems:Experiment Design and Generative Explanation

Powered by advanced information technology,more and more complex systems are exhibiting characteristics of the cyber-physical-social systems(CPSS).In this context,computational experiments method has emerged as a novel approach for the design,analysis,management,control,and integration of CPSS,which can realize the causal analysis of complex systems by means of“algorithmization”of“counterfactuals”.However,because CPSS involve human and social factors(e.g.,autonomy,initiative,and sociality),it is difficult for traditional design of experiment(DOE)methods to achieve the generative explanation of system emergence.To address this challenge,this paper proposes an integrated approach to the design of computational experiments,incorporating three key modules:1)Descriptive module:Determining the influencing factors and response variables of the system by means of the modeling of an artificial society;2)Interpretative module:Selecting factorial experimental design solution to identify the relationship between influencing factors and macro phenomena;3)Predictive module:Building a meta-model that is equivalent to artificial society to explore its operating laws.Finally,a case study of crowd-sourcing platforms is presented to illustrate the application process and effectiveness of the proposed approach,which can reveal the social impact of algorithmic behavior on“rider race”.

General multi-attack detection for continuous-variable quantum key distribution with local local oscillator

Continuous-variable quantum key distribution with a local local oscillator(LLO CVQKD)has been extensively researched due to its simplicity and security.For practical security of an LLO CVQKD system,there are two main attack modes referred to as reference pulse attack and polarization attack presently.However,there is currently no general defense strategy against such attacks,and the security of the system needs further investigation.Here,we employ a deep learning framework called generative adversarial networks(GANs)to detect both attacks.We first analyze the data in different cases,derive a feature vector as input to a GAN model,and then show the training and testing process of the GAN model for attack classification.The proposed model has two parts,a discriminator and a generator,both of which employ a convolutional neural network(CNN)to improve accuracy.Simulation results show that the proposed scheme can detect and classify attacks without reducing the secret key rate and the maximum transmission distance.It only establishes a detection model by monitoring features of the pulse without adding additional devices.

Numerical Study on the Effect of Vortex Generators on the Aerodynamic Drag of a High-Speed Train

A relatively high aerodynamic drag is an important factor that hinders the further acceleration of high-speed trains.Using the shear stress transport(SST)k-ωturbulence model,the effect of various vortex generator types on the aerodynamic characteristics of an ICE2(Inter-city Electricity)train has been investigated.The results indi-cate that the vortex generators with wider triangle,trapezoid,and micro-ramp arranged on the surface of the tail car can significantly change the distribution of surface pressure and affect the vorticity intensity in the wake.This alteration effectively reduces the resistance of the tail car.Meanwhile,the micro-ramp vortex generator with its convergent structure at the rear exhibits enhancedflow-guiding capabilities,resulting in a 15.4%reduction in the drag of the tail car.

A generalization of Banach's lemma and its applications to perturbations of bounded linear operators

Let X be a Banach space and let P:X→X be a bounded linear operator.Using an algebraic inequality on the spectrum of P,we give a new sufficient condition that guarantees the existence of(I-P)^(-1) as a bounded linear operator on X,and a bound on its spectral radius is also obtained.This generalizes the classic Banach lemma.We apply the result to the perturbation analysis of general bounded linear operators on X with commutative perturbations.

Generalized load graphical forecasting method based on modal decomposition

In a“low-carbon”context,the power load is affected by the coupling of multiple factors,which gradually evolves from the traditional“pure load”to the generalized load with the dual characteristics of“load+power supply.”Traditional time-series forecasting methods are no longer suitable owing to the complexity and uncertainty associated with generalized loads.From the perspective of image processing,this study proposes a graphical short-term prediction method for generalized loads based on modal decomposition.First,the datasets are normalized and feature-filtered by comparing the results of Xtreme gradient boosting,gradient boosted decision tree,and random forest algorithms.Subsequently,the generalized load data are decomposed into three sets of modalities by modal decomposition,and red,green,and blue(RGB)images are generated using them as the pixel values of the R,G,and B channels.The generated images are diversified,and an optimized DenseNet neural network was used for training and prediction.Finally,the base load,wind power,and photovoltaic power generation data are selected,and the characteristic curves of the generalized load scenarios under different permeabilities of wind power and photovoltaic power generation are obtained using the density-based spatial clustering of applications with noise algorithm.Based on the proposed graphical forecasting method,the feasibility of the generalized load graphical forecasting method is verified by comparing it with the traditional time-series forecasting method.

Demonstration of a small‐scale power generator using supercritical CO_(2)

The supercritical CO_(2)(sCO_(2))power cycle could improve efficiencies for a wide range of thermal power plants.The sCO_(2)turbine generator plays an important role in the sCO_(2)power cycle by directly converting thermal energy into mechanical work and electric power.The operation of the generator encounters challenges,including high temperature,high pressure,high rotational speed,and other engineering problems,such as leakage.Experimental studies of sCO_(2)turbines are insufficient because of the significant difficulties in turbine manufacturing and system construction.Unlike most experimental investigations that primarily focus on 100 kW‐or MW‐scale power generation systems,we consider,for the first time,a small‐scale power generator using sCO_(2).A partial admission axial turbine was designed and manufactured with a rated rotational speed of 40,000 rpm,and a CO_(2)transcritical power cycle test loop was constructed to validate the performance of our manufactured generator.A resistant gas was proposed in the constructed turbine expander to solve the leakage issue.Both dynamic and steady performances were investigated.The results indicated that a peak electric power of 11.55 kW was achieved at 29,369 rpm.The maximum total efficiency of the turbo‐generator was 58.98%,which was affected by both the turbine rotational speed and pressure ratio,according to the proposed performance map.

Image segmentation of exfoliated two-dimensional materials by generative adversarial network-based data augmentation

Mechanically cleaved two-dimensional materials are random in size and thickness.Recognizing atomically thin flakes by human experts is inefficient and unsuitable for scalable production.Deep learning algorithms have been adopted as an alternative,nevertheless a major challenge is a lack of sufficient actual training images.Here we report the generation of synthetic two-dimensional materials images using StyleGAN3 to complement the dataset.DeepLabv3Plus network is trained with the synthetic images which reduces overfitting and improves recognition accuracy to over 90%.A semi-supervisory technique for labeling images is introduced to reduce manual efforts.The sharper edges recognized by this method facilitate material stacking with precise edge alignment,which benefits exploring novel properties of layered-material devices that crucially depend on the interlayer twist-angle.This feasible and efficient method allows for the rapid and high-quality manufacturing of atomically thin materials and devices.

Multi-distortion suppression for neutron radiographic images based on generative adversarial network

Neutron radiography is a crucial nondestructive testing technology widely used in the aerospace,military,and nuclear industries.However,because of the physical limitations of neutron sources and collimators,the resulting neutron radiographic images inevitably exhibit multiple distortions,including noise,geometric unsharpness,and white spots.Furthermore,these distortions are particularly significant in compact neutron radiography systems with low neutron fluxes.Therefore,in this study,we devised a multi-distortion suppression network that employs a modified generative adversarial network to improve the quality of degraded neutron radiographic images.Real neutron radiographic image datasets with various types and levels of distortion were built for the first time as multi-distortion suppression datasets.Thereafter,the coordinate attention mechanism was incorporated into the backbone network to augment the capability of the proposed network to learn the abstract relationship between ideally clear and degraded images.Extensive experiments were performed;the results show that the proposed method can effectively suppress multiple distortions in real neutron radiographic images and achieve state-of-theart perceptual visual quality,thus demonstrating its application potential in neutron radiography.

Generative adversarial networks based motion learning towards robotic calligraphy synthesis

Robot calligraphy visually reflects the motion capability of robotic manipulators.While traditional researches mainly focus on image generation and the writing of simple calligraphic strokes or characters,this article presents a generative adversarial network(GAN)-based motion learning method for robotic calligraphy synthesis(Gan2CS)that can enhance the efficiency in writing complex calligraphy words and reproducing classic calligraphy works.The key technologies in the proposed approach include:(1)adopting the GAN to learn the motion parameters from the robot writing operation;(2)converting the learnt motion data into the style font and realising the transition from static calligraphy images to dynamic writing demonstration;(3)reproducing high-precision calligraphy works by synthesising the writing motion data hierarchically.In this study,the motion trajectories of sample calligraphy images are firstly extracted and converted into the robot module.The robot performs the writing with motion planning,and the writing motion parameters of calligraphy strokes are learnt with GANs.Then the motion data of basic strokes is synthesised based on the hierarchical process of‘stroke-radicalpart-character’.And the robot re-writes the synthesised characters whose similarity with the original calligraphy characters is evaluated.Regular calligraphy characters have been tested in the experiments for method validation and the results validated that the robot can actualise the robotic calligraphy synthesis of writing motion data with GAN.

Analysis of pseudo-random number generators in QMC-SSE method

In the quantum Monte Carlo(QMC)method,the pseudo-random number generator(PRNG)plays a crucial role in determining the computation time.However,the hidden structure of the PRNG may lead to serious issues such as the breakdown of the Markov process.Here,we systematically analyze the performance of different PRNGs on the widely used QMC method known as the stochastic series expansion(SSE)algorithm.To quantitatively compare them,we introduce a quantity called QMC efficiency that can effectively reflect the efficiency of the algorithms.After testing several representative observables of the Heisenberg model in one and two dimensions,we recommend the linear congruential generator as the best choice of PRNG.Our work not only helps improve the performance of the SSE method but also sheds light on the other Markov-chain-based numerical algorithms.

I-DCGAN and TOPSIS-IFP:A simulation generation model for radiographic flaw detection images in light alloy castings and an algorithm for quality evaluation of generated images

The intelligent detection technology driven by X-ray images and deep learning represents the forefront of advanced techniques and development trends in flaw detection and automated evaluation of light alloy castings.However,the efficacy of deep learning models hinges upon a substantial abundance of flaw samples.The existing research on X-ray image augmentation for flaw detection suffers from shortcomings such as poor diversity of flaw samples and low reliability of quality evaluation.To this end,a novel approach was put forward,which involves the creation of the Interpolation-Deep Convolutional Generative Adversarial Network(I-DCGAN)for flaw detection image generation and a comprehensive evaluation algorithm named TOPSIS-IFP.I-DCGAN enables the generation of high-resolution,diverse simulated images with multiple appearances,achieving an improvement in sample diversity and quality while maintaining a relatively lower computational complexity.TOPSIS-IFP facilitates multi-dimensional quality evaluation,including aspects such as diversity,authenticity,image distribution difference,and image distortion degree.The results indicate that the X-ray radiographic images of magnesium and aluminum alloy castings achieve optimal performance when trained up to the 800th and 600th epochs,respectively.The TOPSIS-IFP value reaches 78.7%and 73.8%similarity to the ideal solution,respectively.Compared to single index evaluation,the TOPSIS-IFP algorithm achieves higher-quality simulated images at the optimal training epoch.This approach successfully mitigates the issue of unreliable quality associated with single index evaluation.The image generation and comprehensive quality evaluation method developed in this paper provides a novel approach for image augmentation in flaw recognition,holding significant importance for enhancing the robustness of subsequent flaw recognition networks.

Conditional Generative Adversarial Network Enabled Localized Stress Recovery of Periodic Composites

Structural damage in heterogeneousmaterials typically originates frommicrostructures where stress concentration occurs.Therefore,evaluating the magnitude and location of localized stress distributions within microstructures under external loading is crucial.Repeating unit cells(RUCs)are commonly used to represent microstructural details and homogenize the effective response of composites.This work develops a machine learning-based micromechanics tool to accurately predict the stress distributions of extracted RUCs.The locally exact homogenization theory efficiently generates the microstructural stresses of RUCs with a wide range of parameters,including volume fraction,fiber/matrix property ratio,fiber shapes,and loading direction.Subsequently,the conditional generative adversarial network(cGAN)is employed and constructed as a surrogate model to establish the statistical correlation between these parameters and the corresponding localized stresses.The stresses predicted by cGAN are validated against the remaining true data not used for training,showing good agreement.This work demonstrates that the cGAN-based micromechanics tool effectively captures the local responses of composite RUCs.It can be used for predicting potential crack initiations starting from microstructures and evaluating the effective behavior of periodic composites.

Bias in Generative AI Systems:A 3-Layer Response and Liability Determination

The risk of bias is widely noticed in the entire process of generative artificial intelligence(generative AI)systems.To protect the rights of the public and improve the effectiveness of AI regulations,feasible measures to address the bias problem in the context of large data should be proposed as soon as possible.Since bias originates in every part and various aspects of AI product lifecycles,laws and technical measures should consider each of these layers and take different causes of bias into account,from data training,modeling,and application design.The Interim Measures for the Administration of Generative AI Service(the Interim Measures),formulated by the Office of the Central Cyberspace Affairs Commission(CAC)and other departments have taken the initiatives to govern AI.However,it lacks specific details on issues such as how to prevent the risk of bias and reduce the effect of bias in decision-making.The Interim Measures also fail to take causes of bias into account,and several principles must be further interpreted.Meanwhile,regulations on generative AI at the global level are still in their early stages.By forming a governance framework,this paper could provide the community with useful experiences and play a leading role.The framework includes at least three parts:first,determining the realm of governance and unifying related concepts;second,developing measures for different layers to identify the causes and specific aspects of bias;third,identifying parties with the skills to take responsibility for detecting bias intrusions and proposing a program for the allocation of liabilities among the large-scale platform developers.

A Novel Defender-Attacker-Defender Model for Resilient Distributed Generator Planning with Network Reconfiguration and Demand Response

To improve the resilience of a distribution system against extreme weather,a fuel-based distributed generator(DG)allocation model is proposed in this study.In this model,the DGs are placed at the planning stage.When an extreme event occurs,the controllable generators form temporary microgrids(MGs)to restore the load maximally.Simultaneously,a demand response program(DRP)mitigates the imbalance between the power supply and demand during extreme events.To cope with the fault uncertainty,a robust optimization(RO)method is applied to reduce the long-term investment and short-term operation costs.The optimization is formulated as a tri-level defenderattacker-defender(DAD)framework.At the first level,decision-makers work out the DG allocation scheme;at the second level,the attacker finds the optimal attack strategy with maximum damage;and at the third level,restoration measures,namely distribution network reconfiguration(DNR)and demand response are performed.The problem is solved by the nested column and constraint generation(NC&CG)method and the model is validated using an IEEE 33-node system.Case studies validate the effectiveness and superiority of the proposed model according to the enhanced resilience and reduced cost.

A Dual Discriminator Method for Generalized Zero-Shot Learning

Zero-shot learning enables the recognition of new class samples by migrating models learned from semanticfeatures and existing sample features to things that have never been seen before. The problems of consistencyof different types of features and domain shift problems are two of the critical issues in zero-shot learning. Toaddress both of these issues, this paper proposes a new modeling structure. The traditional approach mappedsemantic features and visual features into the same feature space;based on this, a dual discriminator approachis used in the proposed model. This dual discriminator approach can further enhance the consistency betweensemantic and visual features. At the same time, this approach can also align unseen class semantic features andtraining set samples, providing a portion of information about the unseen classes. In addition, a new feature fusionmethod is proposed in the model. This method is equivalent to adding perturbation to the seen class features,which can reduce the degree to which the classification results in the model are biased towards the seen classes.At the same time, this feature fusion method can provide part of the information of the unseen classes, improvingits classification accuracy in generalized zero-shot learning and reducing domain bias. The proposed method isvalidated and compared with othermethods on four datasets, and fromthe experimental results, it can be seen thatthe method proposed in this paper achieves promising results.

Quantum generative adversarial networks based on a readout error mitigation method with fault tolerant mechanism

Readout errors caused by measurement noise are a significant source of errors in quantum circuits,which severely affect the output results and are an urgent problem to be solved in noisy-intermediate scale quantum(NISQ)computing.In this paper,we use the bit-flip averaging(BFA)method to mitigate frequent readout errors in quantum generative adversarial networks(QGAN)for image generation,which simplifies the response matrix structure by averaging the qubits for each random bit-flip in advance,successfully solving problems with high cost of measurement for traditional error mitigation methods.Our experiments were simulated in Qiskit using the handwritten digit image recognition dataset under the BFA-based method,the Kullback-Leibler(KL)divergence of the generated images converges to 0.04,0.05,and 0.1 for readout error probabilities of p=0.01,p=0.05,and p=0.1,respectively.Additionally,by evaluating the fidelity of the quantum states representing the images,we observe average fidelity values of 0.97,0.96,and 0.95 for the three readout error probabilities,respectively.These results demonstrate the robustness of the model in mitigating readout errors and provide a highly fault tolerant mechanism for image generation models.

Generative Multi-Modal Mutual Enhancement Video Semantic Communications

Recently,there have been significant advancements in the study of semantic communication in single-modal scenarios.However,the ability to process information in multi-modal environments remains limited.Inspired by the research and applications of natural language processing across different modalities,our goal is to accurately extract frame-level semantic information from videos and ultimately transmit high-quality videos.Specifically,we propose a deep learning-basedMulti-ModalMutual Enhancement Video Semantic Communication system,called M3E-VSC.Built upon a VectorQuantized Generative AdversarialNetwork(VQGAN),our systemaims to leverage mutual enhancement among different modalities by using text as the main carrier of transmission.With it,the semantic information can be extracted fromkey-frame images and audio of the video and performdifferential value to ensure that the extracted text conveys accurate semantic information with fewer bits,thus improving the capacity of the system.Furthermore,a multi-frame semantic detection module is designed to facilitate semantic transitions during video generation.Simulation results demonstrate that our proposed model maintains high robustness in complex noise environments,particularly in low signal-to-noise ratio conditions,significantly improving the accuracy and speed of semantic transmission in video communication by approximately 50 percent.

Electricity Generation from Heatwaves

We chose a definition of heatwaves (HWs) that has ~4-year recurrence frequency at world hot spots. We first examined the 1940-2022 HWs climatology and trends in lifespan, severity, spatial extent, and recurrence frequency. HWs are becoming more frequent and more severe for extratropical mid- and low-latitudes. To euphemize HWs, we here propose a novel clean energy-tapping concept that utilizes the available nano-technology, micro-meteorology knowledge of temperature distribution within/without buildings, and radiative properties of earth atmosphere. The key points for a practical electricity generation scheme from HWs are defogging, insulation, and minimizing the absorption of infrared downward radiation at the cold legs of the thermoelectric generators. One sample realization is presented which, through relay with existing photovoltaic devices, provides all-day electricity supply sufficient for providing air conditioning requirement for a residence (~2000-watt throughput). The provision of power to air conditioning systems, usually imposes a significant stress on traditional city power grids during heatwaves.

Exploring the Implications of the Deformation Parameter and Minimal Length in the Generalized Uncertainty Principle

The breakdown of the Heisenberg Uncertainty Principle occurs when energies approach the Planck scale, and the corresponding Schwarzschild radius becomes similar to the Compton wavelength. Both of these quantities are approximately equal to the Planck length. In this context, we have introduced a model that utilizes a combination of Schwarzschild’s radius and Compton length to quantify the gravitational length of an object. This model has provided a novel perspective in generalizing the uncertainty principle. Furthermore, it has elucidated the significance of the deforming linear parameter β and its range of variation from unity to its maximum value.