Photonic computing has emerged as a promising technology for the ever-increasing computational demands of machine learning and artificial intelligence.Due to the advantages in computing speed,integrated photonic chips...Photonic computing has emerged as a promising technology for the ever-increasing computational demands of machine learning and artificial intelligence.Due to the advantages in computing speed,integrated photonic chips have attracted wide research attention on performing convolutional neural network algorithm.Programmable photonic chips are vital for achieving practical applications of photonic computing.Herein,a programmable photonic chip based on ultrafast laser-induced phase change is fabricated for photonic computing.Through designing the ultrafast laser pulses,the Sb film integrated into photonic waveguides can be reversibly switched between crystalline and amorphous phase,resulting in a large contrast in refractive index and extinction coefficient.As a consequence,the light transmission of waveguides can be switched between write and erase states.To determine the phase change time,the transient laser-induced phase change dynamics of Sb film are revealed at atomic scale,and the time-resolved transient reflectivity is measured.Based on the integrated photonic chip,photonic convolutional neural networks are built to implement machine learning algorithm,and images recognition task is achieved.This work paves a route for fabricating programmable photonic chips by designed ultrafast laser,which will facilitate the application of photonic computing in artificial intelligence.展开更多
The single event effects(SEEs)evaluations caused by atmospheric neutrons were conducted on three different convolutional neural network(CNN)models(Yolov3,MNIST,and ResNet50)in the atmospheric neutron irradiation spect...The single event effects(SEEs)evaluations caused by atmospheric neutrons were conducted on three different convolutional neural network(CNN)models(Yolov3,MNIST,and ResNet50)in the atmospheric neutron irradiation spectrometer(ANIS)at the China Spallation Neutron Source(CSNS).The Yolov3 and MNIST models were implemented on the XILINX28-nm system-on-chip(So C).Meanwhile,the Yolov3 and ResNet50 models were deployed on the XILINX 16-nm Fin FET Ultra Scale+MPSoC.The atmospheric neutron SEEs on the tested CNN systems were comprehensively evaluated from six aspects,including chip type,network architecture,deployment methods,inference time,datasets,and the position of the anchor boxes.The various types of SEE soft errors,SEE cross-sections,and their distribution were analyzed to explore the radiation sensitivities and rules of 28-nm and 16-nm SoC.The current research can provide the technology support of radiation-resistant design of CNN system for developing and applying high-reliability,long-lifespan domestic artificial intelligence chips.展开更多
The shale gas development process is complex in terms of its flow mechanisms and the accuracy of the production forecasting is influenced by geological parameters and engineering parameters.Therefore,to quantitatively...The shale gas development process is complex in terms of its flow mechanisms and the accuracy of the production forecasting is influenced by geological parameters and engineering parameters.Therefore,to quantitatively evaluate the relative importance of model parameters on the production forecasting performance,sensitivity analysis of parameters is required.The parameters are ranked according to the sensitivity coefficients for the subsequent optimization scheme design.A data-driven global sensitivity analysis(GSA)method using convolutional neural networks(CNN)is proposed to identify the influencing parameters in shale gas production.The CNN is trained on a large dataset,validated against numerical simulations,and utilized as a surrogate model for efficient sensitivity analysis.Our approach integrates CNN with the Sobol'global sensitivity analysis method,presenting three key scenarios for sensitivity analysis:analysis of the production stage as a whole,analysis by fixed time intervals,and analysis by declining rate.The findings underscore the predominant influence of reservoir thickness and well length on shale gas production.Furthermore,the temporal sensitivity analysis reveals the dynamic shifts in parameter importance across the distinct production stages.展开更多
High-resolution flow field data has important applications in meteorology,aerospace engineering,high-energy physics and other fields.Experiments and numerical simulations are two main ways to obtain high-resolution fl...High-resolution flow field data has important applications in meteorology,aerospace engineering,high-energy physics and other fields.Experiments and numerical simulations are two main ways to obtain high-resolution flow field data,while the high experiment cost and computing resources for simulation hinder the specificanalysis of flow field evolution.With the development of deep learning technology,convolutional neural networks areused to achieve high-resolution reconstruction of the flow field.In this paper,an ordinary convolutional neuralnetwork and a multi-time-path convolutional neural network are established for the ablative Rayleigh-Taylorinstability.These two methods can reconstruct the high-resolution flow field in just a few seconds,and further greatlyenrich the application of high-resolution reconstruction technology in fluid instability.Compared with the ordinaryconvolutional neural network,the multi-time-path convolutional neural network model has smaller error and canrestore more details of the flow field.The influence of low-resolution flow field data obtained by the two poolingmethods on the convolutional neural networks model is also discussed.展开更多
Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial pro...Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial process parameters and production indicators.While the integrated method of adaptive signal decomposition combined with time series models could effectively predict process variables,it does have limitations in capturing the high-frequency detail of the operation state when applied to complex chemical processes.In light of this,a novel Multiscale Multi-radius Multi-step Convolutional Neural Network(Msrt Net)is proposed for mining spatiotemporal multiscale information.First,the industrial data from the Fluid Catalytic Cracking(FCC)process decomposition using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)extract the multi-energy scale information of the feature subset.Then,convolution kernels with varying stride and padding structures are established to decouple the long-period operation process information encapsulated within the multi-energy scale data.Finally,a reconciliation network is trained to reconstruct the multiscale prediction results and obtain the final output.Msrt Net is initially assessed for its capability to untangle the spatiotemporal multiscale relationships among variables in the Tennessee Eastman Process(TEP).Subsequently,the performance of Msrt Net is evaluated in predicting product yield for a 2.80×10^(6) t/a FCC unit,taking diesel and gasoline yield as examples.In conclusion,Msrt Net can decouple and effectively extract spatiotemporal multiscale information from chemical process data and achieve a approximately reduction of 30%in prediction error compared to other time-series models.Furthermore,its robustness and transferability underscore its promising potential for broader applications.展开更多
We design a new hybrid quantum-classical convolutional neural network(HQCCNN)model based on parameter quantum circuits.In this model,we use parameterized quantum circuits(PQCs)to redesign the convolutional layer in cl...We design a new hybrid quantum-classical convolutional neural network(HQCCNN)model based on parameter quantum circuits.In this model,we use parameterized quantum circuits(PQCs)to redesign the convolutional layer in classical convolutional neural networks,forming a new quantum convolutional layer to achieve unitary transformation of quantum states,enabling the model to more accurately extract hidden information from images.At the same time,we combine the classical fully connected layer with PQCs to form a new hybrid quantum-classical fully connected layer to further improve the accuracy of classification.Finally,we use the MNIST dataset to test the potential of the HQCCNN.The results indicate that the HQCCNN has good performance in solving classification problems.In binary classification tasks,the classification accuracy of numbers 5 and 7 is as high as 99.71%.In multivariate classification,the accuracy rate also reaches 98.51%.Finally,we compare the performance of the HQCCNN with other models and find that the HQCCNN has better classification performance and convergence speed.展开更多
Gas chromatography-mass spectrometry(GC-MS)is an extremely important analytical technique that is widely used in organic geochemistry.It is the only approach to capture biomarker features of organic matter and provide...Gas chromatography-mass spectrometry(GC-MS)is an extremely important analytical technique that is widely used in organic geochemistry.It is the only approach to capture biomarker features of organic matter and provides the key evidence for oil-source correlation and thermal maturity determination.However,the conventional way of processing and interpreting the mass chromatogram is both timeconsuming and labor-intensive,which increases the research cost and restrains extensive applications of this method.To overcome this limitation,a correlation model is developed based on the convolution neural network(CNN)to link the mass chromatogram and biomarker features of samples from the Triassic Yanchang Formation,Ordos Basin,China.In this way,the mass chromatogram can be automatically interpreted.This research first performs dimensionality reduction for 15 biomarker parameters via the factor analysis and then quantifies the biomarker features using two indexes(i.e.MI and PMI)that represent the organic matter thermal maturity and parent material type,respectively.Subsequently,training,interpretation,and validation are performed multiple times using different CNN models to optimize the model structure and hyper-parameter setting,with the mass chromatogram used as the input and the obtained MI and PMI values for supervision(label).The optimized model presents high accuracy in automatically interpreting the mass chromatogram,with R2values typically above 0.85 and0.80 for the thermal maturity and parent material interpretation results,respectively.The significance of this research is twofold:(i)developing an efficient technique for geochemical research;(ii)more importantly,demonstrating the potential of artificial intelligence in organic geochemistry and providing vital references for future related studies.展开更多
Convolutional neural networks(CNNs) exhibit excellent performance in the areas of image recognition and object detection, which can enhance the intelligence level of spacecraft. However, in aerospace, energetic partic...Convolutional neural networks(CNNs) exhibit excellent performance in the areas of image recognition and object detection, which can enhance the intelligence level of spacecraft. However, in aerospace, energetic particles, such as heavy ions, protons, and alpha particles, can induce single event effects(SEEs) that lead CNNs to malfunction and can significantly impact the reliability of a CNN system. In this paper, the MNIST CNN system was constructed based on a 28 nm systemon-chip(SoC), and then an alpha particle irradiation experiment and fault injection were applied to evaluate the SEE of the CNN system. Various types of soft errors in the CNN system have been detected, and the SEE cross sections have been calculated. Furthermore, the mechanisms behind some soft errors have been explained. This research will provide technical support for the design of radiation-resistant artificial intelligence chips.展开更多
Indoor Wi-Fi localization of mobile devices plays a more and more important role along with the rapid growth of location-based services and Wi-Fi mobile devices.In this paper,a new method of constructing the channel s...Indoor Wi-Fi localization of mobile devices plays a more and more important role along with the rapid growth of location-based services and Wi-Fi mobile devices.In this paper,a new method of constructing the channel state information(CSI)image is proposed to improve the localization accuracy.Compared with previous methods of constructing the CSI image,the new kind of CSI image proposed is able to contain more channel information such as the angle of arrival(AoA),the time of arrival(TOA)and the amplitude.We construct three gray images by using phase differences of different antennas and amplitudes of different subcarriers of one antenna,and then merge them to form one RGB image.The localization method has off-line stage and on-line stage.In the off-line stage,the composed three-channel RGB images at training locations are used to train a convolutional neural network(CNN)which has been proved to be efficient in image recognition.In the on-line stage,images at test locations are fed to the well-trained CNN model and the localization result is the weighted mean value with highest output values.The performance of the proposed method is verified with extensive experiments in the representative indoor environment.展开更多
With the continuous progress of The Times and the development of technology,the rise of network social media has also brought the“explosive”growth of image data.As one of the main ways of People’s Daily communicati...With the continuous progress of The Times and the development of technology,the rise of network social media has also brought the“explosive”growth of image data.As one of the main ways of People’s Daily communication,image is widely used as a carrier of communication because of its rich content,intuitive and other advantages.Image recognition based on convolution neural network is the first application in the field of image recognition.A series of algorithm operations such as image eigenvalue extraction,recognition and convolution are used to identify and analyze different images.The rapid development of artificial intelligence makes machine learning more and more important in its research field.Use algorithms to learn each piece of data and predict the outcome.This has become an important key to open the door of artificial intelligence.In machine vision,image recognition is the foundation,but how to associate the low-level information in the image with the high-level image semantics becomes the key problem of image recognition.Predecessors have provided many model algorithms,which have laid a solid foundation for the development of artificial intelligence and image recognition.The multi-level information fusion model based on the VGG16 model is an improvement on the fully connected neural network.Different from full connection network,convolutional neural network does not use full connection method in each layer of neurons of neural network,but USES some nodes for connection.Although this method reduces the computation time,due to the fact that the convolutional neural network model will lose some useful feature information in the process of propagation and calculation,this paper improves the model to be a multi-level information fusion of the convolution calculation method,and further recovers the discarded feature information,so as to improve the recognition rate of the image.VGG divides the network into five groups(mimicking the five layers of AlexNet),yet it USES 3*3 filters and combines them as a convolution sequence.Network deeper DCNN,channel number is bigger.The recognition rate of the model was verified by 0RL Face Database,BioID Face Database and CASIA Face Image Database.展开更多
A method based on multiple images captured under different light sources at different incident angles was developed to recognize the coal density range in this study.The innovation is that two new images were construc...A method based on multiple images captured under different light sources at different incident angles was developed to recognize the coal density range in this study.The innovation is that two new images were constructed based on images captured under four single light sources.Reconstruction image 1 was constructed by fusing greyscale versions of the original images into one image,and Reconstruction image2 was constructed based on the differences between the images captured under the different light sources.Subsequently,the four original images and two reconstructed images were input into the convolutional neural network AlexNet to recognize the density range in three cases:-1.5(clean coal) and+1.5 g/cm^(3)(non-clean coal);-1.8(non-gangue) and+1.8 g/cm^(3)(gangue);-1.5(clean coal),1.5-1.8(middlings),and+1.8 g/cm^(3)(gangue).The results show the following:(1) The reconstructed images,especially Reconstruction image 2,can effectively improve the recognition accuracy for the coal density range compared with images captured under single light source.(2) The recognition accuracies for gangue and non-gangue,clean coal and non-clean coal,and clean coal,middlings,and gangue reached88.44%,86.72% and 77.08%,respectively.(3) The recognition accuracy increases as the density moves further away from the boundary density.展开更多
Neural network methods have recently emerged as a hot topic in computed tomography(CT) imaging owing to their powerful fitting ability;however, their potential applications still need to be carefully studied because t...Neural network methods have recently emerged as a hot topic in computed tomography(CT) imaging owing to their powerful fitting ability;however, their potential applications still need to be carefully studied because their results are often difficult to interpret and are ambiguous in generalizability. Thus, quality assessments of the results obtained from a neural network are necessary to evaluate the neural network. Assessing the image quality of neural networks using traditional objective measurements is not appropriate because neural networks are nonstationary and nonlinear. In contrast, subjective assessments are trustworthy, although they are time-and energy-consuming for radiologists. Model observers that mimic subjective assessment require the mean and covariance of images, which are calculated from numerous image samples;however, this has not yet been applied to the evaluation of neural networks. In this study, we propose an analytical method for noise propagation from a single projection to efficiently evaluate convolutional neural networks(CNNs) in the CT imaging field. We propagate noise through nonlinear layers in a CNN using the Taylor expansion. Nesting of the linear and nonlinear layer noise propagation constitutes the covariance estimation of the CNN. A commonly used U-net structure is adopted for validation. The results reveal that the covariance estimation obtained from the proposed analytical method agrees well with that obtained from the image samples for different phantoms, noise levels, and activation functions, demonstrating that propagating noise from only a single projection is feasible for CNN methods in CT reconstruction. In addition, we use covariance estimation to provide three measurements for the qualitative and quantitative performance evaluation of U-net. The results indicate that the network cannot be applied to projections with high noise levels and possesses limitations in terms of efficiency for processing low-noise projections. U-net is more effective in improving the image quality of smooth regions compared with that of the edge. LeakyReLU outperforms Swish in terms of noise reduction.展开更多
The rapid development of information technology has fueled an ever-increasing demand for ultrafast and ultralow-en-ergy-consumption computing.Existing computing instruments are pre-dominantly electronic processors,whi...The rapid development of information technology has fueled an ever-increasing demand for ultrafast and ultralow-en-ergy-consumption computing.Existing computing instruments are pre-dominantly electronic processors,which use elec-trons as information carriers and possess von Neumann architecture featured by physical separation of storage and pro-cessing.The scaling of computing speed is limited not only by data transfer between memory and processing units,but also by RC delay associated with integrated circuits.Moreover,excessive heating due to Ohmic losses is becoming a severe bottleneck for both speed and power consumption scaling.Using photons as information carriers is a promising alternative.Owing to the weak third-order optical nonlinearity of conventional materials,building integrated photonic com-puting chips under traditional von Neumann architecture has been a challenge.Here,we report a new all-optical comput-ing framework to realize ultrafast and ultralow-energy-consumption all-optical computing based on convolutional neural networks.The device is constructed from cascaded silicon Y-shaped waveguides with side-coupled silicon waveguide segments which we termed“weight modulators”to enable complete phase and amplitude control in each waveguide branch.The generic device concept can be used for equation solving,multifunctional logic operations as well as many other mathematical operations.Multiple computing functions including transcendental equation solvers,multifarious logic gate operators,and half-adders were experimentally demonstrated to validate the all-optical computing performances.The time-of-flight of light through the network structure corresponds to an ultrafast computing time of the order of several picoseconds with an ultralow energy consumption of dozens of femtojoules per bit.Our approach can be further expan-ded to fulfill other complex computing tasks based on non-von Neumann architectures and thus paves a new way for on-chip all-optical computing.展开更多
How to correctly acquire the appropriate features is a primary problem in network protocol recognition field.Aiming to avoid the trouble of artificially extracting features in traditional methods and improve recogniti...How to correctly acquire the appropriate features is a primary problem in network protocol recognition field.Aiming to avoid the trouble of artificially extracting features in traditional methods and improve recognition accuracy,a network protocol recognition method based on Convolutional Neural Network(CNN)is proposed.The method utilizes deep learning technique,and it processes network flows automatically.Firstly,normalization is performed on the intercepted network flows and they are mapped into two-dimensional matrix which will be used as the input of CNN.Then,an improved classification model named Ptr CNN is built,which can automatically extract the appropriate features of network protocols.Finally,the classification model is trained to recognize the network protocols.The proposed approach is compared with several machine learning methods.Experimental results show that the tailored CNN can not only improve protocol recognition accuracy but also ensure the fast convergence of classification model and reduce the classification time.展开更多
A novel convolutional neural network based on spatial pyramid for image classification is proposed.The network exploits image features with spatial pyramid representation.First,it extracts global features from an orig...A novel convolutional neural network based on spatial pyramid for image classification is proposed.The network exploits image features with spatial pyramid representation.First,it extracts global features from an original image,and then different layers of grids are utilized to extract feature maps from different convolutional layers.Inspired by the spatial pyramid,the new network contains two parts,one of which is just like a standard convolutional neural network,composing of alternating convolutions and subsampling layers.But those convolution layers would be averagely pooled by the grid way to obtain feature maps,and then concatenated into a feature vector individually.Finally,those vectors are sequentially concatenated into a total feature vector as the last feature to the fully connection layer.This generated feature vector derives benefits from the classic and previous convolution layer,while the size of the grid adjusting the weight of the feature maps improves the recognition efficiency of the network.Experimental results demonstrate that this model improves the accuracy and applicability compared with the traditional model.展开更多
Traditional fire smoke detection methods mostly rely on manual algorithm extraction and sensor detection;however,these methods are slow and expensive to achieve discrimination.We proposed an improved convolutional neu...Traditional fire smoke detection methods mostly rely on manual algorithm extraction and sensor detection;however,these methods are slow and expensive to achieve discrimination.We proposed an improved convolutional neural network(CNN)to achieve fast analysis.The improved CNN can be used to liberate manpower.The network does not require complicated manual feature extraction to identify forest fire smoke.First,to alleviate the computational pressure and speed up the discrimination efficiency,kernel principal component analysis was performed on the experimental data set.To improve the robustness of the CNN and to avoid overfitting,optimization strategies were applied in multi-convolution kernels and batch normalization to improve loss functions.The experimental analysis shows that the CNN proposed in this study can learn the feature information automatically for smoke images in the early stages of fire automatically with a high recognition rate.As a result,the improved CNN enriches the theory of smoke discrimination in the early stages of a forest fire.展开更多
Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel dat...Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel data-driven framework named convolutional and transformer-based deep neural network(CTDNN)is proposed to improve the classification performance.CTDNN can be divided into four modules,i.e.,convolutional neural network(CNN)backbone,transition module,transformer module,and final classifier.In the CNN backbone,a wide and deep convolution structure is designed,which consists of 1×15 convolution kernels and intensive cross-layer connections instead of traditional 1×3 kernels and sequential connections.In the transition module,a 1×1 convolution layer is utilized to compress the channels of the previous multi-scale CNN features.In the transformer module,three self-attention layers are designed for extracting global features and generating the classification vector.In the classifier,the final decision is made based on the maximum a posterior probability.Extensive simulations are conducted,and the result shows that our proposed CTDNN can achieve superior classification performance than traditional deep models.展开更多
With the rapid growth of the demand for indoor location-based services(LBS), Wi-Fi received signal strength(RSS) fingerprints database has attracted significant attention because it is easy to obtain. The fingerprints...With the rapid growth of the demand for indoor location-based services(LBS), Wi-Fi received signal strength(RSS) fingerprints database has attracted significant attention because it is easy to obtain. The fingerprints algorithm based on convolution neural network(CNN) is often used to improve indoor localization accuracy. However, the number of reference points used for position estimation has significant effects on the positioning accuracy. Meanwhile, it is always selected arbitraily without any guiding standards. As a result, a novel location estimation method based on Jenks natural breaks algorithm(JNBA), which can adaptively choose more reasonable reference points, is proposed in this paper. The output of CNN is processed by JNBA, which can select the number of reference points according to different environments. Then, the location is estimated by weighted K-nearest neighbors(WKNN). Experimental results show that the proposed method has higher positioning accuracy without sacrificing more time cost than the existing indoor localization methods based on CNN.展开更多
To achieve good results in convolutional neural networks(CNN) for text classification task, term-based pooling operation in CNNs is proposed. Firstly, the convolution results of several convolution kernels are combine...To achieve good results in convolutional neural networks(CNN) for text classification task, term-based pooling operation in CNNs is proposed. Firstly, the convolution results of several convolution kernels are combined by this method, and then the results after combination are made pooling operation, three sorts of CNN models(we named TBCNN, MCT-CNN and MMCT-CNN respectively) are constructed and then corresponding algorithmic thought are detailed on this basis. Secondly, relevant experiments and analyses are respectively designed to show the effects of three key parameters(convolution kernel, combination kernel number and word embedding) on three kinds of CNN models and to further demonstrate the effect of the models proposed. The experimental results show that compared with the traditional method of text classification in CNNs, term-based pooling method is addressed that not only the availability of the way is proved, but also the performance shows good superiority.展开更多
Neuromorphic photonic computing has emerged as a competitive computing paradigm to overcome the bottlenecks of the von-Neumann architecture.Linear weighting and nonlinear spike activation are two fundamental functions...Neuromorphic photonic computing has emerged as a competitive computing paradigm to overcome the bottlenecks of the von-Neumann architecture.Linear weighting and nonlinear spike activation are two fundamental functions of a photonic spiking neural network(PSNN).However,they are separately implemented with different photonic materials and devices,hindering the large-scale integration of PSNN.Here,we propose,fabricate and experimentally demonstrate a photonic neuro-synaptic chip enabling the simultaneous implementation of linear weighting and nonlinear spike activation based on a distributed feedback(DFB)laser with a saturable absorber(DFB-SA).A prototypical system is experimentally constructed to demonstrate the parallel weighted function and nonlinear spike activation.Furthermore,a fourchannel DFB-SA laser array is fabricated for realizing matrix convolution of a spiking convolutional neural network,achieving a recognition accuracy of 87%for the MNIST dataset.The fabricated neuro-synaptic chip offers a fundamental building block to construct the large-scale integrated PSNN chip.展开更多
基金supported by the National Key R&D Program of China(2024YFB4609801)the National Natural Science Foundation of China(52075289)the Tsinghua-Jiangyin Innovation Special Fund(TJISF,2023JYTH0104).
文摘Photonic computing has emerged as a promising technology for the ever-increasing computational demands of machine learning and artificial intelligence.Due to the advantages in computing speed,integrated photonic chips have attracted wide research attention on performing convolutional neural network algorithm.Programmable photonic chips are vital for achieving practical applications of photonic computing.Herein,a programmable photonic chip based on ultrafast laser-induced phase change is fabricated for photonic computing.Through designing the ultrafast laser pulses,the Sb film integrated into photonic waveguides can be reversibly switched between crystalline and amorphous phase,resulting in a large contrast in refractive index and extinction coefficient.As a consequence,the light transmission of waveguides can be switched between write and erase states.To determine the phase change time,the transient laser-induced phase change dynamics of Sb film are revealed at atomic scale,and the time-resolved transient reflectivity is measured.Based on the integrated photonic chip,photonic convolutional neural networks are built to implement machine learning algorithm,and images recognition task is achieved.This work paves a route for fabricating programmable photonic chips by designed ultrafast laser,which will facilitate the application of photonic computing in artificial intelligence.
基金Project supported by the National Natural Science Foundation of China(Grant No.12305303)the Natural Science Foundation of Hunan Province of China(Grant Nos.2023JJ40520,2024JJ2044,and 2021JJ40444)+3 种基金the Science and Technology Innovation Program of Hunan Province,China(Grant No.2020RC3054)the Postgraduate Scientific Research Innovation Project of Hunan Province,China(Grant No.CX20240831)the Natural Science Basic Research Plan in the Shaanxi Province of China(Grant No.2023-JC-QN0015)the Doctoral Research Fund of University of South China(Grant No.200XQD033)。
文摘The single event effects(SEEs)evaluations caused by atmospheric neutrons were conducted on three different convolutional neural network(CNN)models(Yolov3,MNIST,and ResNet50)in the atmospheric neutron irradiation spectrometer(ANIS)at the China Spallation Neutron Source(CSNS).The Yolov3 and MNIST models were implemented on the XILINX28-nm system-on-chip(So C).Meanwhile,the Yolov3 and ResNet50 models were deployed on the XILINX 16-nm Fin FET Ultra Scale+MPSoC.The atmospheric neutron SEEs on the tested CNN systems were comprehensively evaluated from six aspects,including chip type,network architecture,deployment methods,inference time,datasets,and the position of the anchor boxes.The various types of SEE soft errors,SEE cross-sections,and their distribution were analyzed to explore the radiation sensitivities and rules of 28-nm and 16-nm SoC.The current research can provide the technology support of radiation-resistant design of CNN system for developing and applying high-reliability,long-lifespan domestic artificial intelligence chips.
基金supported by the National Natural Science Foundation of China (Nos.52274048 and 52374017)Beijing Natural Science Foundation (No.3222037)the CNPC 14th five-year perspective fundamental research project (No.2021DJ2104)。
文摘The shale gas development process is complex in terms of its flow mechanisms and the accuracy of the production forecasting is influenced by geological parameters and engineering parameters.Therefore,to quantitatively evaluate the relative importance of model parameters on the production forecasting performance,sensitivity analysis of parameters is required.The parameters are ranked according to the sensitivity coefficients for the subsequent optimization scheme design.A data-driven global sensitivity analysis(GSA)method using convolutional neural networks(CNN)is proposed to identify the influencing parameters in shale gas production.The CNN is trained on a large dataset,validated against numerical simulations,and utilized as a surrogate model for efficient sensitivity analysis.Our approach integrates CNN with the Sobol'global sensitivity analysis method,presenting three key scenarios for sensitivity analysis:analysis of the production stage as a whole,analysis by fixed time intervals,and analysis by declining rate.The findings underscore the predominant influence of reservoir thickness and well length on shale gas production.Furthermore,the temporal sensitivity analysis reveals the dynamic shifts in parameter importance across the distinct production stages.
基金National Natural Science Foundation of China(1180500311947102+4 种基金12004005)Natural Science Foundation of Anhui Province(2008085MA162008085QA26)University Synergy Innovation Program of Anhui Province(GXXT-2022-039)State Key Laboratory of Advanced Electromagnetic Technology(Grant No.AET 2024KF006)。
文摘High-resolution flow field data has important applications in meteorology,aerospace engineering,high-energy physics and other fields.Experiments and numerical simulations are two main ways to obtain high-resolution flow field data,while the high experiment cost and computing resources for simulation hinder the specificanalysis of flow field evolution.With the development of deep learning technology,convolutional neural networks areused to achieve high-resolution reconstruction of the flow field.In this paper,an ordinary convolutional neuralnetwork and a multi-time-path convolutional neural network are established for the ablative Rayleigh-Taylorinstability.These two methods can reconstruct the high-resolution flow field in just a few seconds,and further greatlyenrich the application of high-resolution reconstruction technology in fluid instability.Compared with the ordinaryconvolutional neural network,the multi-time-path convolutional neural network model has smaller error and canrestore more details of the flow field.The influence of low-resolution flow field data obtained by the two poolingmethods on the convolutional neural networks model is also discussed.
文摘Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial process parameters and production indicators.While the integrated method of adaptive signal decomposition combined with time series models could effectively predict process variables,it does have limitations in capturing the high-frequency detail of the operation state when applied to complex chemical processes.In light of this,a novel Multiscale Multi-radius Multi-step Convolutional Neural Network(Msrt Net)is proposed for mining spatiotemporal multiscale information.First,the industrial data from the Fluid Catalytic Cracking(FCC)process decomposition using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)extract the multi-energy scale information of the feature subset.Then,convolution kernels with varying stride and padding structures are established to decouple the long-period operation process information encapsulated within the multi-energy scale data.Finally,a reconciliation network is trained to reconstruct the multiscale prediction results and obtain the final output.Msrt Net is initially assessed for its capability to untangle the spatiotemporal multiscale relationships among variables in the Tennessee Eastman Process(TEP).Subsequently,the performance of Msrt Net is evaluated in predicting product yield for a 2.80×10^(6) t/a FCC unit,taking diesel and gasoline yield as examples.In conclusion,Msrt Net can decouple and effectively extract spatiotemporal multiscale information from chemical process data and achieve a approximately reduction of 30%in prediction error compared to other time-series models.Furthermore,its robustness and transferability underscore its promising potential for broader applications.
基金Project supported by the Natural Science Foundation of Shandong Province,China (Grant No.ZR2021MF049)the Joint Fund of Natural Science Foundation of Shandong Province (Grant Nos.ZR2022LLZ012 and ZR2021LLZ001)。
文摘We design a new hybrid quantum-classical convolutional neural network(HQCCNN)model based on parameter quantum circuits.In this model,we use parameterized quantum circuits(PQCs)to redesign the convolutional layer in classical convolutional neural networks,forming a new quantum convolutional layer to achieve unitary transformation of quantum states,enabling the model to more accurately extract hidden information from images.At the same time,we combine the classical fully connected layer with PQCs to form a new hybrid quantum-classical fully connected layer to further improve the accuracy of classification.Finally,we use the MNIST dataset to test the potential of the HQCCNN.The results indicate that the HQCCNN has good performance in solving classification problems.In binary classification tasks,the classification accuracy of numbers 5 and 7 is as high as 99.71%.In multivariate classification,the accuracy rate also reaches 98.51%.Finally,we compare the performance of the HQCCNN with other models and find that the HQCCNN has better classification performance and convergence speed.
基金financially supported by China Postdoctoral Science Foundation(Grant No.2023M730365)Natural Science Foundation of Hubei Province of China(Grant No.2023AFB232)。
文摘Gas chromatography-mass spectrometry(GC-MS)is an extremely important analytical technique that is widely used in organic geochemistry.It is the only approach to capture biomarker features of organic matter and provides the key evidence for oil-source correlation and thermal maturity determination.However,the conventional way of processing and interpreting the mass chromatogram is both timeconsuming and labor-intensive,which increases the research cost and restrains extensive applications of this method.To overcome this limitation,a correlation model is developed based on the convolution neural network(CNN)to link the mass chromatogram and biomarker features of samples from the Triassic Yanchang Formation,Ordos Basin,China.In this way,the mass chromatogram can be automatically interpreted.This research first performs dimensionality reduction for 15 biomarker parameters via the factor analysis and then quantifies the biomarker features using two indexes(i.e.MI and PMI)that represent the organic matter thermal maturity and parent material type,respectively.Subsequently,training,interpretation,and validation are performed multiple times using different CNN models to optimize the model structure and hyper-parameter setting,with the mass chromatogram used as the input and the obtained MI and PMI values for supervision(label).The optimized model presents high accuracy in automatically interpreting the mass chromatogram,with R2values typically above 0.85 and0.80 for the thermal maturity and parent material interpretation results,respectively.The significance of this research is twofold:(i)developing an efficient technique for geochemical research;(ii)more importantly,demonstrating the potential of artificial intelligence in organic geochemistry and providing vital references for future related studies.
基金Project supported by the National Natural Science Foundation of China(Grant No.12305303)the Natural Science Foundation of Hunan Province of China(Grant Nos.2023JJ40520,2021JJ40444,and 2019JJ30019)+3 种基金the Research Foundation of Education Bureau of Hunan Province of China(Grant No.20A430)the Science and Technology Innovation Program of Hunan Province(Grant No.2020RC3054)the Natural Science Basic Research Plan in the Shaanxi Province of China(Grant No.2023-JC-QN-0015)the Doctoral Research Fund of University of South China。
文摘Convolutional neural networks(CNNs) exhibit excellent performance in the areas of image recognition and object detection, which can enhance the intelligence level of spacecraft. However, in aerospace, energetic particles, such as heavy ions, protons, and alpha particles, can induce single event effects(SEEs) that lead CNNs to malfunction and can significantly impact the reliability of a CNN system. In this paper, the MNIST CNN system was constructed based on a 28 nm systemon-chip(SoC), and then an alpha particle irradiation experiment and fault injection were applied to evaluate the SEE of the CNN system. Various types of soft errors in the CNN system have been detected, and the SEE cross sections have been calculated. Furthermore, the mechanisms behind some soft errors have been explained. This research will provide technical support for the design of radiation-resistant artificial intelligence chips.
基金supported by the National Natural Science Foundation of China (No.61631013)National Key Basic Research Program of China (973 Program) (No. 2013CB329002)National Major Project (NO. 2018ZX03001006003)
文摘Indoor Wi-Fi localization of mobile devices plays a more and more important role along with the rapid growth of location-based services and Wi-Fi mobile devices.In this paper,a new method of constructing the channel state information(CSI)image is proposed to improve the localization accuracy.Compared with previous methods of constructing the CSI image,the new kind of CSI image proposed is able to contain more channel information such as the angle of arrival(AoA),the time of arrival(TOA)and the amplitude.We construct three gray images by using phase differences of different antennas and amplitudes of different subcarriers of one antenna,and then merge them to form one RGB image.The localization method has off-line stage and on-line stage.In the off-line stage,the composed three-channel RGB images at training locations are used to train a convolutional neural network(CNN)which has been proved to be efficient in image recognition.In the on-line stage,images at test locations are fed to the well-trained CNN model and the localization result is the weighted mean value with highest output values.The performance of the proposed method is verified with extensive experiments in the representative indoor environment.
文摘With the continuous progress of The Times and the development of technology,the rise of network social media has also brought the“explosive”growth of image data.As one of the main ways of People’s Daily communication,image is widely used as a carrier of communication because of its rich content,intuitive and other advantages.Image recognition based on convolution neural network is the first application in the field of image recognition.A series of algorithm operations such as image eigenvalue extraction,recognition and convolution are used to identify and analyze different images.The rapid development of artificial intelligence makes machine learning more and more important in its research field.Use algorithms to learn each piece of data and predict the outcome.This has become an important key to open the door of artificial intelligence.In machine vision,image recognition is the foundation,but how to associate the low-level information in the image with the high-level image semantics becomes the key problem of image recognition.Predecessors have provided many model algorithms,which have laid a solid foundation for the development of artificial intelligence and image recognition.The multi-level information fusion model based on the VGG16 model is an improvement on the fully connected neural network.Different from full connection network,convolutional neural network does not use full connection method in each layer of neurons of neural network,but USES some nodes for connection.Although this method reduces the computation time,due to the fact that the convolutional neural network model will lose some useful feature information in the process of propagation and calculation,this paper improves the model to be a multi-level information fusion of the convolution calculation method,and further recovers the discarded feature information,so as to improve the recognition rate of the image.VGG divides the network into five groups(mimicking the five layers of AlexNet),yet it USES 3*3 filters and combines them as a convolution sequence.Network deeper DCNN,channel number is bigger.The recognition rate of the model was verified by 0RL Face Database,BioID Face Database and CASIA Face Image Database.
文摘A method based on multiple images captured under different light sources at different incident angles was developed to recognize the coal density range in this study.The innovation is that two new images were constructed based on images captured under four single light sources.Reconstruction image 1 was constructed by fusing greyscale versions of the original images into one image,and Reconstruction image2 was constructed based on the differences between the images captured under the different light sources.Subsequently,the four original images and two reconstructed images were input into the convolutional neural network AlexNet to recognize the density range in three cases:-1.5(clean coal) and+1.5 g/cm^(3)(non-clean coal);-1.8(non-gangue) and+1.8 g/cm^(3)(gangue);-1.5(clean coal),1.5-1.8(middlings),and+1.8 g/cm^(3)(gangue).The results show the following:(1) The reconstructed images,especially Reconstruction image 2,can effectively improve the recognition accuracy for the coal density range compared with images captured under single light source.(2) The recognition accuracies for gangue and non-gangue,clean coal and non-clean coal,and clean coal,middlings,and gangue reached88.44%,86.72% and 77.08%,respectively.(3) The recognition accuracy increases as the density moves further away from the boundary density.
基金supported by the National Natural Science Foundation of China(Nos.62031020 and 61771279)。
文摘Neural network methods have recently emerged as a hot topic in computed tomography(CT) imaging owing to their powerful fitting ability;however, their potential applications still need to be carefully studied because their results are often difficult to interpret and are ambiguous in generalizability. Thus, quality assessments of the results obtained from a neural network are necessary to evaluate the neural network. Assessing the image quality of neural networks using traditional objective measurements is not appropriate because neural networks are nonstationary and nonlinear. In contrast, subjective assessments are trustworthy, although they are time-and energy-consuming for radiologists. Model observers that mimic subjective assessment require the mean and covariance of images, which are calculated from numerous image samples;however, this has not yet been applied to the evaluation of neural networks. In this study, we propose an analytical method for noise propagation from a single projection to efficiently evaluate convolutional neural networks(CNNs) in the CT imaging field. We propagate noise through nonlinear layers in a CNN using the Taylor expansion. Nesting of the linear and nonlinear layer noise propagation constitutes the covariance estimation of the CNN. A commonly used U-net structure is adopted for validation. The results reveal that the covariance estimation obtained from the proposed analytical method agrees well with that obtained from the image samples for different phantoms, noise levels, and activation functions, demonstrating that propagating noise from only a single projection is feasible for CNN methods in CT reconstruction. In addition, we use covariance estimation to provide three measurements for the qualitative and quantitative performance evaluation of U-net. The results indicate that the network cannot be applied to projections with high noise levels and possesses limitations in terms of efficiency for processing low-noise projections. U-net is more effective in improving the image quality of smooth regions compared with that of the edge. LeakyReLU outperforms Swish in terms of noise reduction.
基金financial supports from the National Key Research and Development Program of China(2018YFB2200403)National Natural Sci-ence Foundation of China(NSFC)(61775003,11734001,91950204,11527901,11604378,91850117).
文摘The rapid development of information technology has fueled an ever-increasing demand for ultrafast and ultralow-en-ergy-consumption computing.Existing computing instruments are pre-dominantly electronic processors,which use elec-trons as information carriers and possess von Neumann architecture featured by physical separation of storage and pro-cessing.The scaling of computing speed is limited not only by data transfer between memory and processing units,but also by RC delay associated with integrated circuits.Moreover,excessive heating due to Ohmic losses is becoming a severe bottleneck for both speed and power consumption scaling.Using photons as information carriers is a promising alternative.Owing to the weak third-order optical nonlinearity of conventional materials,building integrated photonic com-puting chips under traditional von Neumann architecture has been a challenge.Here,we report a new all-optical comput-ing framework to realize ultrafast and ultralow-energy-consumption all-optical computing based on convolutional neural networks.The device is constructed from cascaded silicon Y-shaped waveguides with side-coupled silicon waveguide segments which we termed“weight modulators”to enable complete phase and amplitude control in each waveguide branch.The generic device concept can be used for equation solving,multifunctional logic operations as well as many other mathematical operations.Multiple computing functions including transcendental equation solvers,multifarious logic gate operators,and half-adders were experimentally demonstrated to validate the all-optical computing performances.The time-of-flight of light through the network structure corresponds to an ultrafast computing time of the order of several picoseconds with an ultralow energy consumption of dozens of femtojoules per bit.Our approach can be further expan-ded to fulfill other complex computing tasks based on non-von Neumann architectures and thus paves a new way for on-chip all-optical computing.
基金supported by the National Key R&D Program of China(2017YFB0802900).
文摘How to correctly acquire the appropriate features is a primary problem in network protocol recognition field.Aiming to avoid the trouble of artificially extracting features in traditional methods and improve recognition accuracy,a network protocol recognition method based on Convolutional Neural Network(CNN)is proposed.The method utilizes deep learning technique,and it processes network flows automatically.Firstly,normalization is performed on the intercepted network flows and they are mapped into two-dimensional matrix which will be used as the input of CNN.Then,an improved classification model named Ptr CNN is built,which can automatically extract the appropriate features of network protocols.Finally,the classification model is trained to recognize the network protocols.The proposed approach is compared with several machine learning methods.Experimental results show that the tailored CNN can not only improve protocol recognition accuracy but also ensure the fast convergence of classification model and reduce the classification time.
基金Supported by the National Natural Science Foundation of China(61601176)the Science and Technology Foundation of Hubei Provincial Department of Education(Q20161405)
文摘A novel convolutional neural network based on spatial pyramid for image classification is proposed.The network exploits image features with spatial pyramid representation.First,it extracts global features from an original image,and then different layers of grids are utilized to extract feature maps from different convolutional layers.Inspired by the spatial pyramid,the new network contains two parts,one of which is just like a standard convolutional neural network,composing of alternating convolutions and subsampling layers.But those convolution layers would be averagely pooled by the grid way to obtain feature maps,and then concatenated into a feature vector individually.Finally,those vectors are sequentially concatenated into a total feature vector as the last feature to the fully connection layer.This generated feature vector derives benefits from the classic and previous convolution layer,while the size of the grid adjusting the weight of the feature maps improves the recognition efficiency of the network.Experimental results demonstrate that this model improves the accuracy and applicability compared with the traditional model.
基金National Natural Science Foundation of China(31670717)Natural Science Foundation of Heilongjiang Province(LH2020C051)。
文摘Traditional fire smoke detection methods mostly rely on manual algorithm extraction and sensor detection;however,these methods are slow and expensive to achieve discrimination.We proposed an improved convolutional neural network(CNN)to achieve fast analysis.The improved CNN can be used to liberate manpower.The network does not require complicated manual feature extraction to identify forest fire smoke.First,to alleviate the computational pressure and speed up the discrimination efficiency,kernel principal component analysis was performed on the experimental data set.To improve the robustness of the CNN and to avoid overfitting,optimization strategies were applied in multi-convolution kernels and batch normalization to improve loss functions.The experimental analysis shows that the CNN proposed in this study can learn the feature information automatically for smoke images in the early stages of fire automatically with a high recognition rate.As a result,the improved CNN enriches the theory of smoke discrimination in the early stages of a forest fire.
基金supported in part by the National Natural Science Foundation of China under Grant(62171045,62201090)in part by the National Key Research and Development Program of China under Grants(2020YFB1807602,2019YFB1804404).
文摘Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel data-driven framework named convolutional and transformer-based deep neural network(CTDNN)is proposed to improve the classification performance.CTDNN can be divided into four modules,i.e.,convolutional neural network(CNN)backbone,transition module,transformer module,and final classifier.In the CNN backbone,a wide and deep convolution structure is designed,which consists of 1×15 convolution kernels and intensive cross-layer connections instead of traditional 1×3 kernels and sequential connections.In the transition module,a 1×1 convolution layer is utilized to compress the channels of the previous multi-scale CNN features.In the transformer module,three self-attention layers are designed for extracting global features and generating the classification vector.In the classifier,the final decision is made based on the maximum a posterior probability.Extensive simulations are conducted,and the result shows that our proposed CTDNN can achieve superior classification performance than traditional deep models.
基金supported by the National Natural Science Foundation of China (NSFC) under Grants 62001238 and 61901075。
文摘With the rapid growth of the demand for indoor location-based services(LBS), Wi-Fi received signal strength(RSS) fingerprints database has attracted significant attention because it is easy to obtain. The fingerprints algorithm based on convolution neural network(CNN) is often used to improve indoor localization accuracy. However, the number of reference points used for position estimation has significant effects on the positioning accuracy. Meanwhile, it is always selected arbitraily without any guiding standards. As a result, a novel location estimation method based on Jenks natural breaks algorithm(JNBA), which can adaptively choose more reasonable reference points, is proposed in this paper. The output of CNN is processed by JNBA, which can select the number of reference points according to different environments. Then, the location is estimated by weighted K-nearest neighbors(WKNN). Experimental results show that the proposed method has higher positioning accuracy without sacrificing more time cost than the existing indoor localization methods based on CNN.
文摘To achieve good results in convolutional neural networks(CNN) for text classification task, term-based pooling operation in CNNs is proposed. Firstly, the convolution results of several convolution kernels are combined by this method, and then the results after combination are made pooling operation, three sorts of CNN models(we named TBCNN, MCT-CNN and MMCT-CNN respectively) are constructed and then corresponding algorithmic thought are detailed on this basis. Secondly, relevant experiments and analyses are respectively designed to show the effects of three key parameters(convolution kernel, combination kernel number and word embedding) on three kinds of CNN models and to further demonstrate the effect of the models proposed. The experimental results show that compared with the traditional method of text classification in CNNs, term-based pooling method is addressed that not only the availability of the way is proved, but also the performance shows good superiority.
基金financial supports from National Key Research and Development Program of China (2021YFB2801900,2021YFB2801901,2021YFB2801902,2021YFB2801904)National Natural Science Foundation of China (No.61974177)+1 种基金National Outstanding Youth Science Fund Project of National Natural Science Foundation of China (62022062)The Fundamental Research Funds for the Central Universities (QTZX23041).
文摘Neuromorphic photonic computing has emerged as a competitive computing paradigm to overcome the bottlenecks of the von-Neumann architecture.Linear weighting and nonlinear spike activation are two fundamental functions of a photonic spiking neural network(PSNN).However,they are separately implemented with different photonic materials and devices,hindering the large-scale integration of PSNN.Here,we propose,fabricate and experimentally demonstrate a photonic neuro-synaptic chip enabling the simultaneous implementation of linear weighting and nonlinear spike activation based on a distributed feedback(DFB)laser with a saturable absorber(DFB-SA).A prototypical system is experimentally constructed to demonstrate the parallel weighted function and nonlinear spike activation.Furthermore,a fourchannel DFB-SA laser array is fabricated for realizing matrix convolution of a spiking convolutional neural network,achieving a recognition accuracy of 87%for the MNIST dataset.The fabricated neuro-synaptic chip offers a fundamental building block to construct the large-scale integrated PSNN chip.
