The thermal conductivity of nanofluids is an important property that influences the heat transfer capabilities of nanofluids.Researchers rely on experimental investigations to explore nanofluid properties,as it is a n...The thermal conductivity of nanofluids is an important property that influences the heat transfer capabilities of nanofluids.Researchers rely on experimental investigations to explore nanofluid properties,as it is a necessary step before their practical application.As these investigations are time and resource-consuming undertakings,an effective prediction model can significantly improve the efficiency of research operations.In this work,an Artificial Neural Network(ANN)model is developed to predict the thermal conductivity of metal oxide water-based nanofluid.For this,a comprehensive set of 691 data points was collected from the literature.This dataset is split into training(70%),validation(15%),and testing(15%)and used to train the ANN model.The developed model is a backpropagation artificial neural network with a 4–12–1 architecture.The performance of the developed model shows high accuracy with R values above 0.90 and rapid convergence.It shows that the developed ANN model accurately predicts the thermal conductivity of nanofluids.展开更多
Skin diseases affect millions worldwide.Early detection is key to preventing disfigurement,lifelong disability,or death.Dermoscopic images acquired in primary-care settings show high intra-class visual similarity and ...Skin diseases affect millions worldwide.Early detection is key to preventing disfigurement,lifelong disability,or death.Dermoscopic images acquired in primary-care settings show high intra-class visual similarity and severe class imbalance,and occasional imaging artifacts can create ambiguity for state-of-the-art convolutional neural networks(CNNs).We frame skin lesion recognition as graph-based reasoning and,to ensure fair evaluation and avoid data leakage,adopt a strict lesion-level partitioning strategy.Each image is first over-segmented using SLIC(Simple Linear Iterative Clustering)to produce perceptually homogeneous superpixels.These superpixels form the nodes of a region-adjacency graph whose edges encode spatial continuity.Node attributes are 1280-dimensional embeddings extracted with a lightweight yet expressive EfficientNet-B0 backbone,providing strong representational power at modest computational cost.The resulting graphs are processed by a five-layer Graph Attention Network(GAT)that learns to weight inter-node relationships dynamically and aggregates multi-hop context before classifying lesions into seven classes with a log-softmax output.Extensive experiments on the DermaMNIST benchmark show the proposed pipeline achieves 88.35%accuracy and 98.04%AUC,outperforming contemporary CNNs,AutoML approaches,and alternative graph neural networks.An ablation study indicates EfficientNet-B0 produces superior node descriptors compared with ResNet-18 and DenseNet,and that roughly five GAT layers strike a good balance between being too shallow and over-deep while avoiding oversmoothing.The method requires no data augmentation or external metadata,making it a drop-in upgrade for clinical computer-aided diagnosis systems.展开更多
With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performan...With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.展开更多
The increasing popularity of the Internet and the widespread use of information technology have led to a rise in the number and sophistication of network attacks and security threats.Intrusion detection systems are cr...The increasing popularity of the Internet and the widespread use of information technology have led to a rise in the number and sophistication of network attacks and security threats.Intrusion detection systems are crucial to network security,playing a pivotal role in safeguarding networks from potential threats.However,in the context of an evolving landscape of sophisticated and elusive attacks,existing intrusion detection methodologies often overlook critical aspects such as changes in network topology over time and interactions between hosts.To address these issues,this paper proposes a real-time network intrusion detection method based on graph neural networks.The proposedmethod leverages the advantages of graph neural networks and employs a straightforward graph construction method to represent network traffic as dynamic graph-structured data.Additionally,a graph convolution operation with a multi-head attention mechanism is utilized to enhance the model’s ability to capture the intricate relationships within the graph structure comprehensively.Furthermore,it uses an integrated graph neural network to address dynamic graphs’structural and topological changes at different time points and the challenges of edge embedding in intrusion detection data.The edge classification problem is effectively transformed into node classification by employing a line graph data representation,which facilitates fine-grained intrusion detection tasks on dynamic graph node feature representations.The efficacy of the proposed method is evaluated using two commonly used intrusion detection datasets,UNSW-NB15 and NF-ToN-IoT-v2,and results are compared with previous studies in this field.The experimental results demonstrate that our proposed method achieves 99.3%and 99.96%accuracy on the two datasets,respectively,and outperforms the benchmark model in several evaluation metrics.展开更多
Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using d...Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.展开更多
Real-time prediction and precise control of sinter quality are pivotal for energy saving,cost reduction,quality improvement and efficiency enhancement in the ironmaking process.To advance,the accuracy and comprehensiv...Real-time prediction and precise control of sinter quality are pivotal for energy saving,cost reduction,quality improvement and efficiency enhancement in the ironmaking process.To advance,the accuracy and comprehensiveness of sinter quality prediction,an intelligent flare monitoring system for sintering machine tails that combines hybrid neural networks integrating convolutional neural network with long short-term memory(CNN-LSTM)networks was proposed.The system utilized a high-temperature thermal imager for image acquisition at the sintering machine tail and employed a zone-triggered method to accurately capture dynamic feature images under challenging conditions of high-temperature,high dust,and occlusion.The feature images were then segmented through a triple-iteration multi-thresholding approach based on the maximum between-class variance method to minimize detail loss during the segmentation process.Leveraging the advantages of CNN and LSTM networks in capturing temporal and spatial information,a comprehensive model for sinter quality prediction was constructed,with inputs including the proportion of combustion layer,porosity rate,temperature distribution,and image features obtained from the convolutional neural network,and outputs comprising quality indicators such as underburning index,uniformity index,and FeO content of the sinter.The accuracy is notably increased,achieving a 95.8%hit rate within an error margin of±1.0.After the system is applied,the average qualified rate of FeO content increases from 87.24%to 89.99%,representing an improvement of 2.75%.The average monthly solid fuel consumption is reduced from 49.75 to 46.44 kg/t,leading to a 6.65%reduction and underscoring significant energy saving and cost reduction effects.展开更多
Micro-expressions(ME)recognition is a complex task that requires advanced techniques to extract informative features fromfacial expressions.Numerous deep neural networks(DNNs)with convolutional structures have been pr...Micro-expressions(ME)recognition is a complex task that requires advanced techniques to extract informative features fromfacial expressions.Numerous deep neural networks(DNNs)with convolutional structures have been proposed.However,unlike DNNs,shallow convolutional neural networks often outperform deeper models in mitigating overfitting,particularly with small datasets.Still,many of these methods rely on a single feature for recognition,resulting in an insufficient ability to extract highly effective features.To address this limitation,in this paper,an Improved Dual-stream Shallow Convolutional Neural Network based on an Extreme Gradient Boosting Algorithm(IDSSCNN-XgBoost)is introduced for ME Recognition.The proposed method utilizes a dual-stream architecture where motion vectors(temporal features)are extracted using Optical Flow TV-L1 and amplify subtle changes(spatial features)via EulerianVideoMagnification(EVM).These features are processed by IDSSCNN,with an attention mechanism applied to refine the extracted effective features.The outputs are then fused,concatenated,and classified using the XgBoost algorithm.This comprehensive approach significantly improves recognition accuracy by leveraging the strengths of both temporal and spatial information,supported by the robust classification power of XgBoost.The proposed method is evaluated on three publicly available ME databases named Chinese Academy of Sciences Micro-expression Database(CASMEII),Spontaneous Micro-Expression Database(SMICHS),and Spontaneous Actions and Micro-Movements(SAMM).Experimental results indicate that the proposed model can achieve outstanding results compared to recent models.The accuracy results are 79.01%,69.22%,and 68.99%on CASMEII,SMIC-HS,and SAMM,and the F1-score are 75.47%,68.91%,and 63.84%,respectively.The proposed method has the advantage of operational efficiency and less computational time.展开更多
Spiking Neural Network(SNN)inspired by the biological triggering mechanism of neurons to provide a novel solution for plant disease detection,offering enhanced performance and efficiency in contrast to Artificial Neur...Spiking Neural Network(SNN)inspired by the biological triggering mechanism of neurons to provide a novel solution for plant disease detection,offering enhanced performance and efficiency in contrast to Artificial Neural Networks(ANN).Unlike conventional ANNs,which process static images without fully capturing the inherent temporal dynamics,our approach represents the first implementation of SNNs tailored explicitly for agricultural disease classification,integrating an encoding method to convert static RGB plant images into temporally encoded spike trains.Additionally,while Bernoulli trials and standard deep learning architectures likeConvolutionalNeuralNetworks(CNNs)and Fully Connected Neural Networks(FCNNs)have been used extensively,our work is the first to integrate these trials within an SNN framework specifically for agricultural applications.This integration not only refines spike regulation and reduces computational overhead by 30%but also delivers superior accuracy(93.4%)in plant disease classification,marking a significant advancement in precision agriculture that has not been previously explored.Our approach uniquely transforms static plant leaf images into time-dependent representations,leveraging SNNs’intrinsic temporal processing capabilities.This approach aligns with the inherent ability of SNNs to capture dynamic,timedependent patterns,making them more suitable for detecting disease activations in plants than conventional ANNs that treat inputs as static entities.Unlike prior works,our hybrid encoding scheme dynamically adapts to pixel intensity variations(via threshold),enabling robust feature extraction under diverse agricultural conditions.The dual-stage preprocessing customizes the SNN’s behavior in two ways:the encoding threshold is derived from pixel distributions in diseased regions,and Bernoulli trials selectively reduce redundant spikes to ensure energy efficiency on low-power devices.We used a comprehensive dataset of 87,000 RGB images of plant leaves,which included 38 distinct classes of healthy and unhealthy leaves.To train and evaluate three distinct neural network architectures,DeepSNN,SimpleCNN,and SimpleFCNN,the dataset was rigorously preprocessed,including stochastic rotation,horizontal flip,resizing,and normalization.Moreover,by integrating Bernoulli trials to regulate spike generation,ourmethod focuses on extracting themost relevant featureswhile reducingcomputational overhead.Using a comprehensivedatasetof87,000RGB images across 38 classes,we rigorously preprocessed the data and evaluated three architectures:DeepSNN,SimpleCNN,and SimpleFCNN.The results demonstrate that DeepSNN outperforms the other models,achieving superior accuracy,efficient feature extraction,and robust spike management,thereby establishing the potential of SNNs for real-time,energy-efficient agricultural applications.展开更多
Blockchain platform swith the unique characteristics of anonymity,decentralization,and transparency of their transactions,which are faced with abnormal activities such as money laundering,phishing scams,and fraudulent...Blockchain platform swith the unique characteristics of anonymity,decentralization,and transparency of their transactions,which are faced with abnormal activities such as money laundering,phishing scams,and fraudulent behavior,posing a serious threat to account asset security.For these potential security risks,this paper proposes a hybrid neural network detection method(HNND)that learns multiple types of account features and enhances fusion information among them to effectively detect abnormal transaction behaviors in the blockchain.In HNND,the Temporal Transaction Graph Attention Network(T2GAT)is first designed to learn biased aggregation representation of multi-attribute transactions among nodes,which can capture key temporal information from node neighborhood transactions.Then,the Graph Convolutional Network(GCN)is adopted which captures abstract structural features of the transaction network.Further,the Stacked Denoising Autoencode(SDA)is developed to achieve adaptive fusion of thses features from different modules.Moreover,the SDA enhances robustness and generalization ability of node representation,leading to higher binary classification accuracy in detecting abnormal behaviors of blockchain accounts.Evaluations on a real-world abnormal transaction dataset demonstrate great advantages of the proposed HNND method over other compared methods.展开更多
Underwater acoustic target recognition(UATR)has become increasingly prevalent for ocean detection,localisation,and identification.However,due to the complexity and variability of underwater environments,especially in ...Underwater acoustic target recognition(UATR)has become increasingly prevalent for ocean detection,localisation,and identification.However,due to the complexity and variability of underwater environments,especially in multi ship event environments,where multiple acoustic signals coexist,practical applications face significant challenges.These challenges hinder single-category acoustic recognition algorithms,particularly in extracting time series features and achieving fine-grained or multi-scale feature fusion.This paper innovatively introduce the SKANN framework,which achieve precise submarine sound recognition in underwater mixed ship events environments through timing data enhancement and sampling training module and selective kernel feature extraction module.The timing data enhancement and sampling training module improves time sequence feature extraction through progressive acoustic sampling.The selective kernel feature extraction module effectively fuses multi-scale features by integrating selective kernel(SK)technology.To simulate concurrent ship events,we constructed the mixed ship noise dataset(MDeepShip),providing an experimental basis and test platform for underwater mixed ship event detection.This dataset ensures that the model encounters diverse audio samples during training and validation,improving its ability to extract temporal features.Experimental results show that SKANN achieves a 93.6%recognition rate on the M-DeepShip dataset,demonstrating its effectiveness in recognising underwater mixed ship events.Given the complexity of real underwater environments,this work lays a crucial foundation for the sound recognition of submarine vessels.Future research will focus on real marine environments to validate and refine the models and methods for practical applications.展开更多
BACKGROUND Rumination is a critical psychological factor contributing to the relapse of major depressive episodes(MDEs)and a core residual symptom in remitted MDEs.Investigating its neural correlations is essential fo...BACKGROUND Rumination is a critical psychological factor contributing to the relapse of major depressive episodes(MDEs)and a core residual symptom in remitted MDEs.Investigating its neural correlations is essential for developing strategies to prevent MDE relapse.Despite its clinical importance,the brain network mechanisms underlying rumination in remitted MDE patients have yet to be fully elucidated.AIM To investigate the brain network mechanism underlying rumination in patients with remitted MDEs using functional magnetic resonance imaging(fMRI).METHODS We conducted an fMRI-based rumination-distraction task to induce rumination and distraction states in 51 patients with remitted MDEs.Functional connectivity(FC)was analyzed using the network-based statistic(NBS)approach,and eight topological metrics were calculated to compare the network topological properties between the two states.Correlation analyses were further performed to identify the relationships between individual rumination levels and the significantly altered brain network metrics.RESULTS The NBS analysis revealed that the altered FCs between the rumination and distraction states were located primarily in the frontoparietal,default mode,and cerebellar networks.No significant correlation was detected between these altered FCs and individual rumination levels.Among the eight topological metrics,the clustering coefficient,shortest path length,and local efficiency were significantly lower during rumination and positively correlated with individual rumination levels.In contrast,global efficiency was greater in the rumination state than in the distraction state and was negatively correlated with individual rumination levels.CONCLUSION Our work revealed the altered FC and topological properties during rumination in remitted MDE patients,offering valuable insights into the neural mechanisms of rumination from a brain network perspective.展开更多
Artificial intelligence(AI)is a revolutionizing problem-solver across various domains,including scientific research.Its application to chemical processes holds remarkable potential for rapid optimization of protocols ...Artificial intelligence(AI)is a revolutionizing problem-solver across various domains,including scientific research.Its application to chemical processes holds remarkable potential for rapid optimization of protocols and methods.A notable application of AI is in the photoFenton degradation of organic compounds.Despite the high novelty and recent surge of interest in this area,a comprehensive synthesis of existing literature on AI applications in the photo-Fenton process is lacking.This review aims to bridge this gap by providing an in-depth summary of the state-of-the-art use of artificial neural networks(ANN)in the photo-Fenton process,with the goal of aiding researchers in the water treatment field to identify the most crucial and relevant variables.It examines the types and architectures of ANNs,input and output variables,and the efficiency of these networks.The findings reveal a rapidly expanding field with increasing publications highlighting AI's potential to optimize the photo-Fenton process.This review also discusses the benefits and drawbacks of using ANNs,emphasizing the need for further research to advance this promising area.展开更多
Solar chimneys are renewable energy systems designed to enhance natural ventilation,improving thermal comfort in buildings.As passive systems,solar chimneys contribute to energy efficiency in a sustainable and environ...Solar chimneys are renewable energy systems designed to enhance natural ventilation,improving thermal comfort in buildings.As passive systems,solar chimneys contribute to energy efficiency in a sustainable and environmentally friendly way.The effectiveness of a solar chimney depends on its design and orientation relative to the cardinal directions,both of which are critical for optimal performance.This article presents a supervised learning approach using artificial neural networks to forecast the performance indicators of solar chimneys.Thedataset includes information from 2784 solar chimney configurations,which encompasses various factors such as chimney height,channel thickness,glass thickness,paint,wall material,measurement date,and orientation.The case study examines the four cardinal orientations and weather data from Mexico City,covering the period from 01 January to 31 December 2024.The main results indicate that the proposed artificial neural network models achieved higher coefficient of determination values(0.905-0.990)than the baseline method across performance indicators of the solar chimney system,demonstrating greater accuracy and improved generalization.The proposed approach highlights the potential of using artificial neural networks as a decision-making tool in the design stage of solar chimneys in sustainable architecture.展开更多
Graph Neural Networks(GNNs)have demonstrated outstanding capabilities in processing graph-structured data and are increasingly being integrated into large-scale pre-trained models,such as Large Language Models(LLMs),t...Graph Neural Networks(GNNs)have demonstrated outstanding capabilities in processing graph-structured data and are increasingly being integrated into large-scale pre-trained models,such as Large Language Models(LLMs),to enhance structural reasoning,knowledge retrieval,and memory management.The expansion of their application scope imposes higher requirements on the robustness of GNNs.However,as GNNs are applied to more dynamic and heterogeneous environments,they become increasingly vulnerable to real-world perturbations.In particular,graph data frequently encounters joint adversarial perturbations that simultaneously affect both structures and features,which are significantly more challenging than isolated attacks.These disruptions,caused by incomplete data,malicious attacks,or inherent noise,pose substantial threats to the stable and reliable performance of traditional GNN models.To address this issue,this study proposes the Dual-Shield Graph Neural Network(DSGNN),a defense model that simultaneously mitigates structural and feature perturbations.DSGNN utilizes two parallel GNN channels to independently process structural noise and feature noise,and introduces an adaptive fusion mechanism that integrates information from both pathways to generate robust node representations.Theoretical analysis demonstrates that DSGNN achieves a tighter robustness boundary under joint perturbations compared to conventional single-channel methods.Experimental evaluations across Cora,CiteSeer,and Industry datasets show that DSGNN achieves the highest average classification accuracy under various adversarial settings,reaching 81.24%,71.94%,and 81.66%,respectively,outperforming GNNGuard,GCN-Jaccard,GCN-SVD,RGCN,and NoisyGNN.These results underscore the importance of multi-view perturbation decoupling in constructing resilient GNN models for real-world applications.展开更多
Controlling the directionality of quantum emitter(QE)radiation is crucial for advancing nanophotonic devices,yet designing compact,single nanoantennas for this purpose remains challenging with traditional methods due ...Controlling the directionality of quantum emitter(QE)radiation is crucial for advancing nanophotonic devices,yet designing compact,single nanoantennas for this purpose remains challenging with traditional methods due to computational demands and optimization complexity.This paper introduces a neural-network-based inverse design approach to efficiently optimize single core-shell nanosphere antennas and their derivatives—notched core-shell structures and dimers—for highly directional QE emission.We systematically compare the radial basis function(RBF),support vector regression(SVR),and backpropagation neural network(BPNN)algorithms.Our results demonstrate BPNN's superior performance in accurately mapping nanoantennas'geometric and material parameters to their far-field radiation characteristics.Subsequent BPNN-driven optimization confirms that all three investigated structures(single core-shell,notched core-shell,and dimer)can achieve robust directional emission.This is accomplished by precisely exciting higher-order electric and magnetic multipoles engineered to possess equal amplitudes and opposite phases,thereby facilitating directional radiation from QEs and enabling more efficient nanophotonic device design.展开更多
Physics-informed neural networks(PINNs)have shown remarkable prospects in solving the forward and inverse problems involving partial differential equations(PDEs).The method embeds PDEs into the neural network by calcu...Physics-informed neural networks(PINNs)have shown remarkable prospects in solving the forward and inverse problems involving partial differential equations(PDEs).The method embeds PDEs into the neural network by calculating the PDE loss at a set of collocation points,providing advantages such as meshfree and more convenient adaptive sampling.However,when solving PDEs using nonuniform collocation points,PINNs still face challenge regarding inefficient convergence of PDE residuals or even failure.In this work,we first analyze the ill-conditioning of the PDE loss in PINNs under nonuniform collocation points.To address the issue,we define volume weighting residual and propose volume weighting physics-informed neural networks(VW-PINNs).Through weighting the PDE residuals by the volume that the collocation points occupy within the computational domain,we embed explicitly the distribution characteristics of collocation points in the loss evaluation.The fast and sufficient convergence of the PDE residuals for the problems involving nonuniform collocation points is guaranteed.Considering the meshfree characteristics of VW-PINNs,we also develop a volume approximation algorithm based on kernel density estimation to calculate the volume of the collocation points.We validate the universality of VW-PINNs by solving the forward problems involving flow over a circular cylinder and flow over the NACA0012 airfoil under different inflow conditions,where conventional PINNs fail.By solving the Burgers’equation,we verify that VW-PINNs can enhance the efficiency of existing the adaptive sampling method in solving the forward problem by three times,and can reduce the relative L 2 error of conventional PINNs in solving the inverse problem by more than one order of magnitude.展开更多
Federated Graph Neural Networks (FedGNNs) have achieved significant success in representation learning for graph data, enabling collaborative training among multiple parties without sharing their raw graph data and so...Federated Graph Neural Networks (FedGNNs) have achieved significant success in representation learning for graph data, enabling collaborative training among multiple parties without sharing their raw graph data and solving the data isolation problem faced by centralized GNNs in data-sensitive scenarios. Despite the plethora of prior work on inference attacks against centralized GNNs, the vulnerability of FedGNNs to inference attacks has not yet been widely explored. It is still unclear whether the privacy leakage risks of centralized GNNs will also be introduced in FedGNNs. To bridge this gap, we present PIAFGNN, the first property inference attack (PIA) against FedGNNs. Compared with prior works on centralized GNNs, in PIAFGNN, the attacker can only obtain the global embedding gradient distributed by the central server. The attacker converts the task of stealing the target user’s local embeddings into a regression problem, using a regression model to generate the target graph node embeddings. By training shadow models and property classifiers, the attacker can infer the basic property information within the target graph that is of interest. Experiments on three benchmark graph datasets demonstrate that PIAFGNN achieves attack accuracy of over 70% in most cases, even approaching the attack accuracy of inference attacks against centralized GNNs in some instances, which is much higher than the attack accuracy of the random guessing method. Furthermore, we observe that common defense mechanisms cannot mitigate our attack without affecting the model’s performance on mainly classification tasks.展开更多
The quantum hybrid algorithm has recently become a very promising and speedy method for solving larger-scale optimization problems in the noisy intermediate-scale quantum(NISQ)era.The unit commitment(UC)problem is a f...The quantum hybrid algorithm has recently become a very promising and speedy method for solving larger-scale optimization problems in the noisy intermediate-scale quantum(NISQ)era.The unit commitment(UC)problem is a fundamental problem in the field of power systems that aims to satisfy the power balance constraint with minimal cost.In this paper,we focus on the implementation of the UC solution using exact quantum algorithms based on the quantum neural network(QNN).This method is tested with a ten-unit system under the power balance constraint.In order to improve computing precision and reduce network complexity,we propose a knowledge-based partially connected quantum neural network(PCQNN).The results show that exact solutions can be obtained by the improved algorithm and that the depth of the quantum circuit can be reduced simultaneously.展开更多
Physics-informed neural networks(PINNs)have shown considerable promise for performing numerical simulations in fluid mechanics.They provide mesh-free,end-to-end approaches by embedding physical laws into their loss fu...Physics-informed neural networks(PINNs)have shown considerable promise for performing numerical simulations in fluid mechanics.They provide mesh-free,end-to-end approaches by embedding physical laws into their loss functions.However,when addressing complex flow problems,PINNs still face some challenges such as activation saturation and vanishing gradients in deep network training,leading to slow convergence and insufficient prediction accuracy.We present physics-informed neural networks incorporating lattice Boltzmann method optimized by tanh robust weight initialization(T-PINN-LBM)to address these challenges.This approach fuses the mesoscopic lattice Boltzmann model with the automatic differentiation framework of PINNs.It also implements a tanh robust weight initialization method derived from fixed point analysis.This model effectively mitigates activation and gradient decay in deep networks,improving convergence speed and data efficiency in multiscale flow simulations.We validate the effectiveness of the model on the classical arithmetic example of lid-driven cavity flow.Compared to the traditional Xavier initialized PINN and PINN-LBM,T-PINNLBM reduces the mean absolute error(MAE)by one order of magnitude at the same network depth and maintains stable convergence in deeper networks.The results demonstrate that this model can accurately capture complex flow structures without prior data,providing a new feasible pathway for data-free driven fluid simulation.展开更多
Due to their biological interpretability,memristors are widely used to simulate synapses between artificial neural networks.As a type of neural network whose dynamic behavior can be explained,the coupling of resonant ...Due to their biological interpretability,memristors are widely used to simulate synapses between artificial neural networks.As a type of neural network whose dynamic behavior can be explained,the coupling of resonant tunneling diode-based cellular neural networks(RTD-CNNs)with memristors has rarely been reported in the literature.Therefore,this paper designs a coupled RTD-CNN model with memristors(RTD-MCNN),investigating and analyzing the dynamic behavior of the RTD-MCNN.Based on this model,a simple encryption scheme for the protection of digital images in police forensic applications is proposed.The results show that the RTD-MCNN can have two positive Lyapunov exponents,and its output is influenced by the initial values,exhibiting multistability.Furthermore,a set of amplitudes in its output sequence is affected by the internal parameters of the memristor,leading to nonlinear variations.Undoubtedly,the rich dynamic behaviors described above make the RTD-MCNN highly suitable for the design of chaos-based encryption schemes in the field of privacy protection.Encryption tests and security analyses validate the effectiveness of this scheme.展开更多
基金supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2021R1A6A1A10044950).
文摘The thermal conductivity of nanofluids is an important property that influences the heat transfer capabilities of nanofluids.Researchers rely on experimental investigations to explore nanofluid properties,as it is a necessary step before their practical application.As these investigations are time and resource-consuming undertakings,an effective prediction model can significantly improve the efficiency of research operations.In this work,an Artificial Neural Network(ANN)model is developed to predict the thermal conductivity of metal oxide water-based nanofluid.For this,a comprehensive set of 691 data points was collected from the literature.This dataset is split into training(70%),validation(15%),and testing(15%)and used to train the ANN model.The developed model is a backpropagation artificial neural network with a 4–12–1 architecture.The performance of the developed model shows high accuracy with R values above 0.90 and rapid convergence.It shows that the developed ANN model accurately predicts the thermal conductivity of nanofluids.
基金funded by the Deanship of Graduate Studies and Scientific Research at Jouf University under grant No.(DGSSR-2025-02-01296).
文摘Skin diseases affect millions worldwide.Early detection is key to preventing disfigurement,lifelong disability,or death.Dermoscopic images acquired in primary-care settings show high intra-class visual similarity and severe class imbalance,and occasional imaging artifacts can create ambiguity for state-of-the-art convolutional neural networks(CNNs).We frame skin lesion recognition as graph-based reasoning and,to ensure fair evaluation and avoid data leakage,adopt a strict lesion-level partitioning strategy.Each image is first over-segmented using SLIC(Simple Linear Iterative Clustering)to produce perceptually homogeneous superpixels.These superpixels form the nodes of a region-adjacency graph whose edges encode spatial continuity.Node attributes are 1280-dimensional embeddings extracted with a lightweight yet expressive EfficientNet-B0 backbone,providing strong representational power at modest computational cost.The resulting graphs are processed by a five-layer Graph Attention Network(GAT)that learns to weight inter-node relationships dynamically and aggregates multi-hop context before classifying lesions into seven classes with a log-softmax output.Extensive experiments on the DermaMNIST benchmark show the proposed pipeline achieves 88.35%accuracy and 98.04%AUC,outperforming contemporary CNNs,AutoML approaches,and alternative graph neural networks.An ablation study indicates EfficientNet-B0 produces superior node descriptors compared with ResNet-18 and DenseNet,and that roughly five GAT layers strike a good balance between being too shallow and over-deep while avoiding oversmoothing.The method requires no data augmentation or external metadata,making it a drop-in upgrade for clinical computer-aided diagnosis systems.
基金supported by the Natural Science Foundation of China No.62362008the Major Scientific and Technological Special Project of Guizhou Province([2024]014).
文摘With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.
文摘The increasing popularity of the Internet and the widespread use of information technology have led to a rise in the number and sophistication of network attacks and security threats.Intrusion detection systems are crucial to network security,playing a pivotal role in safeguarding networks from potential threats.However,in the context of an evolving landscape of sophisticated and elusive attacks,existing intrusion detection methodologies often overlook critical aspects such as changes in network topology over time and interactions between hosts.To address these issues,this paper proposes a real-time network intrusion detection method based on graph neural networks.The proposedmethod leverages the advantages of graph neural networks and employs a straightforward graph construction method to represent network traffic as dynamic graph-structured data.Additionally,a graph convolution operation with a multi-head attention mechanism is utilized to enhance the model’s ability to capture the intricate relationships within the graph structure comprehensively.Furthermore,it uses an integrated graph neural network to address dynamic graphs’structural and topological changes at different time points and the challenges of edge embedding in intrusion detection data.The edge classification problem is effectively transformed into node classification by employing a line graph data representation,which facilitates fine-grained intrusion detection tasks on dynamic graph node feature representations.The efficacy of the proposed method is evaluated using two commonly used intrusion detection datasets,UNSW-NB15 and NF-ToN-IoT-v2,and results are compared with previous studies in this field.The experimental results demonstrate that our proposed method achieves 99.3%and 99.96%accuracy on the two datasets,respectively,and outperforms the benchmark model in several evaluation metrics.
基金The work described in this paper was fully supported by a grant from Hong Kong Metropolitan University(RIF/2021/05).
文摘Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.
基金founded by the Open Project Program of Anhui Province Key Laboratory of Metallurgical Engineering and Resources Recycling(Anhui University of Technology)(No.SKF21-06)Research Fund for Young Teachers of Anhui University of Technology in 2020(No.QZ202001).
文摘Real-time prediction and precise control of sinter quality are pivotal for energy saving,cost reduction,quality improvement and efficiency enhancement in the ironmaking process.To advance,the accuracy and comprehensiveness of sinter quality prediction,an intelligent flare monitoring system for sintering machine tails that combines hybrid neural networks integrating convolutional neural network with long short-term memory(CNN-LSTM)networks was proposed.The system utilized a high-temperature thermal imager for image acquisition at the sintering machine tail and employed a zone-triggered method to accurately capture dynamic feature images under challenging conditions of high-temperature,high dust,and occlusion.The feature images were then segmented through a triple-iteration multi-thresholding approach based on the maximum between-class variance method to minimize detail loss during the segmentation process.Leveraging the advantages of CNN and LSTM networks in capturing temporal and spatial information,a comprehensive model for sinter quality prediction was constructed,with inputs including the proportion of combustion layer,porosity rate,temperature distribution,and image features obtained from the convolutional neural network,and outputs comprising quality indicators such as underburning index,uniformity index,and FeO content of the sinter.The accuracy is notably increased,achieving a 95.8%hit rate within an error margin of±1.0.After the system is applied,the average qualified rate of FeO content increases from 87.24%to 89.99%,representing an improvement of 2.75%.The average monthly solid fuel consumption is reduced from 49.75 to 46.44 kg/t,leading to a 6.65%reduction and underscoring significant energy saving and cost reduction effects.
基金supported by the Key Research and Development Program of Jiangsu Province under Grant BE2022059-3,CTBC Bank through the Industry-Academia Cooperation Project,as well as by the Ministry of Science and Technology of Taiwan through Grants MOST-108-2218-E-002-055,MOST-109-2223-E-009-002-MY3,MOST-109-2218-E-009-025,and MOST431109-2218-E-002-015.
文摘Micro-expressions(ME)recognition is a complex task that requires advanced techniques to extract informative features fromfacial expressions.Numerous deep neural networks(DNNs)with convolutional structures have been proposed.However,unlike DNNs,shallow convolutional neural networks often outperform deeper models in mitigating overfitting,particularly with small datasets.Still,many of these methods rely on a single feature for recognition,resulting in an insufficient ability to extract highly effective features.To address this limitation,in this paper,an Improved Dual-stream Shallow Convolutional Neural Network based on an Extreme Gradient Boosting Algorithm(IDSSCNN-XgBoost)is introduced for ME Recognition.The proposed method utilizes a dual-stream architecture where motion vectors(temporal features)are extracted using Optical Flow TV-L1 and amplify subtle changes(spatial features)via EulerianVideoMagnification(EVM).These features are processed by IDSSCNN,with an attention mechanism applied to refine the extracted effective features.The outputs are then fused,concatenated,and classified using the XgBoost algorithm.This comprehensive approach significantly improves recognition accuracy by leveraging the strengths of both temporal and spatial information,supported by the robust classification power of XgBoost.The proposed method is evaluated on three publicly available ME databases named Chinese Academy of Sciences Micro-expression Database(CASMEII),Spontaneous Micro-Expression Database(SMICHS),and Spontaneous Actions and Micro-Movements(SAMM).Experimental results indicate that the proposed model can achieve outstanding results compared to recent models.The accuracy results are 79.01%,69.22%,and 68.99%on CASMEII,SMIC-HS,and SAMM,and the F1-score are 75.47%,68.91%,and 63.84%,respectively.The proposed method has the advantage of operational efficiency and less computational time.
基金supported in part by the Basic Science Research Program through the National Research Foundation of Korea(NRF),funded by the Ministry of Education(NRF-2021R1A6A1A03039493).
文摘Spiking Neural Network(SNN)inspired by the biological triggering mechanism of neurons to provide a novel solution for plant disease detection,offering enhanced performance and efficiency in contrast to Artificial Neural Networks(ANN).Unlike conventional ANNs,which process static images without fully capturing the inherent temporal dynamics,our approach represents the first implementation of SNNs tailored explicitly for agricultural disease classification,integrating an encoding method to convert static RGB plant images into temporally encoded spike trains.Additionally,while Bernoulli trials and standard deep learning architectures likeConvolutionalNeuralNetworks(CNNs)and Fully Connected Neural Networks(FCNNs)have been used extensively,our work is the first to integrate these trials within an SNN framework specifically for agricultural applications.This integration not only refines spike regulation and reduces computational overhead by 30%but also delivers superior accuracy(93.4%)in plant disease classification,marking a significant advancement in precision agriculture that has not been previously explored.Our approach uniquely transforms static plant leaf images into time-dependent representations,leveraging SNNs’intrinsic temporal processing capabilities.This approach aligns with the inherent ability of SNNs to capture dynamic,timedependent patterns,making them more suitable for detecting disease activations in plants than conventional ANNs that treat inputs as static entities.Unlike prior works,our hybrid encoding scheme dynamically adapts to pixel intensity variations(via threshold),enabling robust feature extraction under diverse agricultural conditions.The dual-stage preprocessing customizes the SNN’s behavior in two ways:the encoding threshold is derived from pixel distributions in diseased regions,and Bernoulli trials selectively reduce redundant spikes to ensure energy efficiency on low-power devices.We used a comprehensive dataset of 87,000 RGB images of plant leaves,which included 38 distinct classes of healthy and unhealthy leaves.To train and evaluate three distinct neural network architectures,DeepSNN,SimpleCNN,and SimpleFCNN,the dataset was rigorously preprocessed,including stochastic rotation,horizontal flip,resizing,and normalization.Moreover,by integrating Bernoulli trials to regulate spike generation,ourmethod focuses on extracting themost relevant featureswhile reducingcomputational overhead.Using a comprehensivedatasetof87,000RGB images across 38 classes,we rigorously preprocessed the data and evaluated three architectures:DeepSNN,SimpleCNN,and SimpleFCNN.The results demonstrate that DeepSNN outperforms the other models,achieving superior accuracy,efficient feature extraction,and robust spike management,thereby establishing the potential of SNNs for real-time,energy-efficient agricultural applications.
文摘Blockchain platform swith the unique characteristics of anonymity,decentralization,and transparency of their transactions,which are faced with abnormal activities such as money laundering,phishing scams,and fraudulent behavior,posing a serious threat to account asset security.For these potential security risks,this paper proposes a hybrid neural network detection method(HNND)that learns multiple types of account features and enhances fusion information among them to effectively detect abnormal transaction behaviors in the blockchain.In HNND,the Temporal Transaction Graph Attention Network(T2GAT)is first designed to learn biased aggregation representation of multi-attribute transactions among nodes,which can capture key temporal information from node neighborhood transactions.Then,the Graph Convolutional Network(GCN)is adopted which captures abstract structural features of the transaction network.Further,the Stacked Denoising Autoencode(SDA)is developed to achieve adaptive fusion of thses features from different modules.Moreover,the SDA enhances robustness and generalization ability of node representation,leading to higher binary classification accuracy in detecting abnormal behaviors of blockchain accounts.Evaluations on a real-world abnormal transaction dataset demonstrate great advantages of the proposed HNND method over other compared methods.
基金funded by The National Natural Science Foundation of China under Grant(Nos.62273108 and 62306081)The Youth Project of Guangdong Artificial Intelligence and Digital Economy Laboratory(Guangzhou)(PZL2022KF0006)+6 种基金The National Key Research and Development Program-Research on Key technology of High Frequency broadband mobile communication credit Filter and its Industrialization application-Subproject Circuit Design and Simulation of high frequency broadband Filter(2022YFB3604502)‘New Generation Information Technology’Major Science and Technology Project of Guangzhou Key Field R&D Plan(202206070001)the Special Fund Project of Guangzhou Science and Technology Innovation Development(202201011307)the Guangdong Provincial Department of Education Key construction discipline Scientific research ability Improvement Project,Introduction of Talents Project of Guangdong Polytechnic Normal University of China(99166990222)the Special Projects in Key Fields of General Colleges and Universities in Guangdong Province(2021ZDZX1016)the Natural Science Foundation of Guangdong Province(2024A1515010120)the Special Fund Project of Guangzhou Science and Technology Innovation Development(202201011307).
文摘Underwater acoustic target recognition(UATR)has become increasingly prevalent for ocean detection,localisation,and identification.However,due to the complexity and variability of underwater environments,especially in multi ship event environments,where multiple acoustic signals coexist,practical applications face significant challenges.These challenges hinder single-category acoustic recognition algorithms,particularly in extracting time series features and achieving fine-grained or multi-scale feature fusion.This paper innovatively introduce the SKANN framework,which achieve precise submarine sound recognition in underwater mixed ship events environments through timing data enhancement and sampling training module and selective kernel feature extraction module.The timing data enhancement and sampling training module improves time sequence feature extraction through progressive acoustic sampling.The selective kernel feature extraction module effectively fuses multi-scale features by integrating selective kernel(SK)technology.To simulate concurrent ship events,we constructed the mixed ship noise dataset(MDeepShip),providing an experimental basis and test platform for underwater mixed ship event detection.This dataset ensures that the model encounters diverse audio samples during training and validation,improving its ability to extract temporal features.Experimental results show that SKANN achieves a 93.6%recognition rate on the M-DeepShip dataset,demonstrating its effectiveness in recognising underwater mixed ship events.Given the complexity of real underwater environments,this work lays a crucial foundation for the sound recognition of submarine vessels.Future research will focus on real marine environments to validate and refine the models and methods for practical applications.
基金the National Key Research and Development Program of China,No.2021ZD0202000the National Natural Science Foundation of China,No.82101612 and No.82471570+1 种基金the Natural Science Foundation of Hunan Province,China,No.2022JJ40692the Science and Technology Innovation Program of Hunan Province,No.2021RC2040 and No.2024RC3056.
文摘BACKGROUND Rumination is a critical psychological factor contributing to the relapse of major depressive episodes(MDEs)and a core residual symptom in remitted MDEs.Investigating its neural correlations is essential for developing strategies to prevent MDE relapse.Despite its clinical importance,the brain network mechanisms underlying rumination in remitted MDE patients have yet to be fully elucidated.AIM To investigate the brain network mechanism underlying rumination in patients with remitted MDEs using functional magnetic resonance imaging(fMRI).METHODS We conducted an fMRI-based rumination-distraction task to induce rumination and distraction states in 51 patients with remitted MDEs.Functional connectivity(FC)was analyzed using the network-based statistic(NBS)approach,and eight topological metrics were calculated to compare the network topological properties between the two states.Correlation analyses were further performed to identify the relationships between individual rumination levels and the significantly altered brain network metrics.RESULTS The NBS analysis revealed that the altered FCs between the rumination and distraction states were located primarily in the frontoparietal,default mode,and cerebellar networks.No significant correlation was detected between these altered FCs and individual rumination levels.Among the eight topological metrics,the clustering coefficient,shortest path length,and local efficiency were significantly lower during rumination and positively correlated with individual rumination levels.In contrast,global efficiency was greater in the rumination state than in the distraction state and was negatively correlated with individual rumination levels.CONCLUSION Our work revealed the altered FC and topological properties during rumination in remitted MDE patients,offering valuable insights into the neural mechanisms of rumination from a brain network perspective.
基金financial support provided by the Valencian Regional Governement(Grant No.CIPROM2023/037)Davide Palma and Alessandra Bianco Prevot acknowledge support from the Project CH4.0 under the MUR program"Dipartimenti di Eccellenza 2023-2027"(Grant No.CUP:D13C22003520001).
文摘Artificial intelligence(AI)is a revolutionizing problem-solver across various domains,including scientific research.Its application to chemical processes holds remarkable potential for rapid optimization of protocols and methods.A notable application of AI is in the photoFenton degradation of organic compounds.Despite the high novelty and recent surge of interest in this area,a comprehensive synthesis of existing literature on AI applications in the photo-Fenton process is lacking.This review aims to bridge this gap by providing an in-depth summary of the state-of-the-art use of artificial neural networks(ANN)in the photo-Fenton process,with the goal of aiding researchers in the water treatment field to identify the most crucial and relevant variables.It examines the types and architectures of ANNs,input and output variables,and the efficiency of these networks.The findings reveal a rapidly expanding field with increasing publications highlighting AI's potential to optimize the photo-Fenton process.This review also discusses the benefits and drawbacks of using ANNs,emphasizing the need for further research to advance this promising area.
文摘Solar chimneys are renewable energy systems designed to enhance natural ventilation,improving thermal comfort in buildings.As passive systems,solar chimneys contribute to energy efficiency in a sustainable and environmentally friendly way.The effectiveness of a solar chimney depends on its design and orientation relative to the cardinal directions,both of which are critical for optimal performance.This article presents a supervised learning approach using artificial neural networks to forecast the performance indicators of solar chimneys.Thedataset includes information from 2784 solar chimney configurations,which encompasses various factors such as chimney height,channel thickness,glass thickness,paint,wall material,measurement date,and orientation.The case study examines the four cardinal orientations and weather data from Mexico City,covering the period from 01 January to 31 December 2024.The main results indicate that the proposed artificial neural network models achieved higher coefficient of determination values(0.905-0.990)than the baseline method across performance indicators of the solar chimney system,demonstrating greater accuracy and improved generalization.The proposed approach highlights the potential of using artificial neural networks as a decision-making tool in the design stage of solar chimneys in sustainable architecture.
基金funded by the Key Research and Development Program of Zhejiang Province No.2023C01141the Science and Technology Innovation Community Project of the Yangtze River Delta No.23002410100suported by the Open Research Fund of the State Key Laboratory of Blockchain and Data Security,Zhejiang University.
文摘Graph Neural Networks(GNNs)have demonstrated outstanding capabilities in processing graph-structured data and are increasingly being integrated into large-scale pre-trained models,such as Large Language Models(LLMs),to enhance structural reasoning,knowledge retrieval,and memory management.The expansion of their application scope imposes higher requirements on the robustness of GNNs.However,as GNNs are applied to more dynamic and heterogeneous environments,they become increasingly vulnerable to real-world perturbations.In particular,graph data frequently encounters joint adversarial perturbations that simultaneously affect both structures and features,which are significantly more challenging than isolated attacks.These disruptions,caused by incomplete data,malicious attacks,or inherent noise,pose substantial threats to the stable and reliable performance of traditional GNN models.To address this issue,this study proposes the Dual-Shield Graph Neural Network(DSGNN),a defense model that simultaneously mitigates structural and feature perturbations.DSGNN utilizes two parallel GNN channels to independently process structural noise and feature noise,and introduces an adaptive fusion mechanism that integrates information from both pathways to generate robust node representations.Theoretical analysis demonstrates that DSGNN achieves a tighter robustness boundary under joint perturbations compared to conventional single-channel methods.Experimental evaluations across Cora,CiteSeer,and Industry datasets show that DSGNN achieves the highest average classification accuracy under various adversarial settings,reaching 81.24%,71.94%,and 81.66%,respectively,outperforming GNNGuard,GCN-Jaccard,GCN-SVD,RGCN,and NoisyGNN.These results underscore the importance of multi-view perturbation decoupling in constructing resilient GNN models for real-world applications.
基金Project supported by the Key Research Projects of Colleges and Universities in Anhui Province(Grant No.KJ2021A0492)the Pre-research Project of the National Natural Science Foundation,An Hui Polytechnic University(Grant Nos.Xjky2022042 and Xjky2022048)+2 种基金the Testing Technology and Energy Saving Device Anhui Provincial Laboratory Open Fund(Grant No.JCKJ2022A09)the Joint Opening Project of Anhui Engineering Research Center of Vehicle Display Integrated Systems and Joint Discipline Key Laboratory of Touch Display Materials and Devices in Anhui Province(Grant No.VDIS2023B02)the Science and Technology Project of Wuhu(Grant No.2023jc05)。
文摘Controlling the directionality of quantum emitter(QE)radiation is crucial for advancing nanophotonic devices,yet designing compact,single nanoantennas for this purpose remains challenging with traditional methods due to computational demands and optimization complexity.This paper introduces a neural-network-based inverse design approach to efficiently optimize single core-shell nanosphere antennas and their derivatives—notched core-shell structures and dimers—for highly directional QE emission.We systematically compare the radial basis function(RBF),support vector regression(SVR),and backpropagation neural network(BPNN)algorithms.Our results demonstrate BPNN's superior performance in accurately mapping nanoantennas'geometric and material parameters to their far-field radiation characteristics.Subsequent BPNN-driven optimization confirms that all three investigated structures(single core-shell,notched core-shell,and dimer)can achieve robust directional emission.This is accomplished by precisely exciting higher-order electric and magnetic multipoles engineered to possess equal amplitudes and opposite phases,thereby facilitating directional radiation from QEs and enabling more efficient nanophotonic device design.
基金supported by the National Natural Science Foundation of China(Grant No.92152301)the National Key Research and Development Program of China(Grant No.2022YFB4300200)the Shaanxi Provincial Key Research and Development Program(Grant No.2023-ZDLGY-27).
文摘Physics-informed neural networks(PINNs)have shown remarkable prospects in solving the forward and inverse problems involving partial differential equations(PDEs).The method embeds PDEs into the neural network by calculating the PDE loss at a set of collocation points,providing advantages such as meshfree and more convenient adaptive sampling.However,when solving PDEs using nonuniform collocation points,PINNs still face challenge regarding inefficient convergence of PDE residuals or even failure.In this work,we first analyze the ill-conditioning of the PDE loss in PINNs under nonuniform collocation points.To address the issue,we define volume weighting residual and propose volume weighting physics-informed neural networks(VW-PINNs).Through weighting the PDE residuals by the volume that the collocation points occupy within the computational domain,we embed explicitly the distribution characteristics of collocation points in the loss evaluation.The fast and sufficient convergence of the PDE residuals for the problems involving nonuniform collocation points is guaranteed.Considering the meshfree characteristics of VW-PINNs,we also develop a volume approximation algorithm based on kernel density estimation to calculate the volume of the collocation points.We validate the universality of VW-PINNs by solving the forward problems involving flow over a circular cylinder and flow over the NACA0012 airfoil under different inflow conditions,where conventional PINNs fail.By solving the Burgers’equation,we verify that VW-PINNs can enhance the efficiency of existing the adaptive sampling method in solving the forward problem by three times,and can reduce the relative L 2 error of conventional PINNs in solving the inverse problem by more than one order of magnitude.
基金supported by the National Natural Science Foundation of China(Nos.62176122 and 62061146002).
文摘Federated Graph Neural Networks (FedGNNs) have achieved significant success in representation learning for graph data, enabling collaborative training among multiple parties without sharing their raw graph data and solving the data isolation problem faced by centralized GNNs in data-sensitive scenarios. Despite the plethora of prior work on inference attacks against centralized GNNs, the vulnerability of FedGNNs to inference attacks has not yet been widely explored. It is still unclear whether the privacy leakage risks of centralized GNNs will also be introduced in FedGNNs. To bridge this gap, we present PIAFGNN, the first property inference attack (PIA) against FedGNNs. Compared with prior works on centralized GNNs, in PIAFGNN, the attacker can only obtain the global embedding gradient distributed by the central server. The attacker converts the task of stealing the target user’s local embeddings into a regression problem, using a regression model to generate the target graph node embeddings. By training shadow models and property classifiers, the attacker can infer the basic property information within the target graph that is of interest. Experiments on three benchmark graph datasets demonstrate that PIAFGNN achieves attack accuracy of over 70% in most cases, even approaching the attack accuracy of inference attacks against centralized GNNs in some instances, which is much higher than the attack accuracy of the random guessing method. Furthermore, we observe that common defense mechanisms cannot mitigate our attack without affecting the model’s performance on mainly classification tasks.
基金supported in part by the China Postdoctoral Science Foundation(Grant No.2023M740874)。
文摘The quantum hybrid algorithm has recently become a very promising and speedy method for solving larger-scale optimization problems in the noisy intermediate-scale quantum(NISQ)era.The unit commitment(UC)problem is a fundamental problem in the field of power systems that aims to satisfy the power balance constraint with minimal cost.In this paper,we focus on the implementation of the UC solution using exact quantum algorithms based on the quantum neural network(QNN).This method is tested with a ten-unit system under the power balance constraint.In order to improve computing precision and reduce network complexity,we propose a knowledge-based partially connected quantum neural network(PCQNN).The results show that exact solutions can be obtained by the improved algorithm and that the depth of the quantum circuit can be reduced simultaneously.
基金supported by the National Natural Science Foundation of China(Grant Nos.12474453,12174085,and 12404530).
文摘Physics-informed neural networks(PINNs)have shown considerable promise for performing numerical simulations in fluid mechanics.They provide mesh-free,end-to-end approaches by embedding physical laws into their loss functions.However,when addressing complex flow problems,PINNs still face some challenges such as activation saturation and vanishing gradients in deep network training,leading to slow convergence and insufficient prediction accuracy.We present physics-informed neural networks incorporating lattice Boltzmann method optimized by tanh robust weight initialization(T-PINN-LBM)to address these challenges.This approach fuses the mesoscopic lattice Boltzmann model with the automatic differentiation framework of PINNs.It also implements a tanh robust weight initialization method derived from fixed point analysis.This model effectively mitigates activation and gradient decay in deep networks,improving convergence speed and data efficiency in multiscale flow simulations.We validate the effectiveness of the model on the classical arithmetic example of lid-driven cavity flow.Compared to the traditional Xavier initialized PINN and PINN-LBM,T-PINNLBM reduces the mean absolute error(MAE)by one order of magnitude at the same network depth and maintains stable convergence in deeper networks.The results demonstrate that this model can accurately capture complex flow structures without prior data,providing a new feasible pathway for data-free driven fluid simulation.
基金supported by the Scientific Research Fund of Hunan Provincial Education Department(Grant No.24A0248)the National Key Research and Development Program“National Quality Infrastructure System”Special Project(Grant No.2024YFF0617900)the Hefei Minglong Electronic Technology Co.,Ltd.(Grant Nos.2024ZKHX293,2024ZKHX294,and 2024ZKHX295).
文摘Due to their biological interpretability,memristors are widely used to simulate synapses between artificial neural networks.As a type of neural network whose dynamic behavior can be explained,the coupling of resonant tunneling diode-based cellular neural networks(RTD-CNNs)with memristors has rarely been reported in the literature.Therefore,this paper designs a coupled RTD-CNN model with memristors(RTD-MCNN),investigating and analyzing the dynamic behavior of the RTD-MCNN.Based on this model,a simple encryption scheme for the protection of digital images in police forensic applications is proposed.The results show that the RTD-MCNN can have two positive Lyapunov exponents,and its output is influenced by the initial values,exhibiting multistability.Furthermore,a set of amplitudes in its output sequence is affected by the internal parameters of the memristor,leading to nonlinear variations.Undoubtedly,the rich dynamic behaviors described above make the RTD-MCNN highly suitable for the design of chaos-based encryption schemes in the field of privacy protection.Encryption tests and security analyses validate the effectiveness of this scheme.
