Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Re...Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.展开更多
Objective To address the dual challenges of long-tail distribution and feature sparsity in traditional Chinese medicine(TCM)syndrome differentiation within real clinical settings,we propose a data-efficient learning f...Objective To address the dual challenges of long-tail distribution and feature sparsity in traditional Chinese medicine(TCM)syndrome differentiation within real clinical settings,we propose a data-efficient learning framework enhanced by knowledge graphs.Methods We developed Agent-GNN,a three-stage decoupled learning framework,and validated it on the Traditional Chinese Medicine Syndrome Diagnosis(TCM-SD)dataset containing 54152 clinical records across 148 syndrome categories.First,we constructed a comprehensive medical knowledge graph encoding the complete TCM reasoning system.Second,we proposed a Functional Patient Profiling(FPP)method that utilizes large language models(LLMs)combined with Graph Retrieval-Augmented Generation(RAG)to extract structured symptom-etiology-pathogenesis subgraphs from medical records.Third,we employed heterogeneous graph neural networks to learn structured combination patterns explicitly.We compared our method against multiple baselines including BERT,ZY-BERT,ZY-BERT+Know,GAT,and GPT-4 Few-shot,using macro-F1 score as the primary evaluation metric.Additionally,ablation experiments were conducted to validate the contribution of each key component to model performance.Results Agent-GNN achieved an overall macro-F1 score of 72.4%,representing an 8.7 percentage points improvement over ZY-BERT+Know(63.7%),the strongest baseline among traditional methods.For long-tail syndromes with fewer than 10 samples,Agent-GNN reached a macro-F1 score of 58.6%,compared with 39.3%for ZY-BERT+Know and 41.2%for GPT-4 Few-shot,representing relative improvements of 49.2%and 42.2%,respectively.Ablation experiments confirmed that the explicit modeling of etiology-pathogenesis nodes contributed 12.4 percentage points to this enhanced long-tail syndrome performance.Conclusion This study proposes Agent-GNN,a knowledge graph-enhanced framework that effectively addresses the long-tail distribution challenge in TCM syndrome differentiation.By explicitly modeling manifestation-mechanism-essence patterns through structured knowledge graphs,our approach achieves superior performance in data-scarce scenarios while providing interpretable reasoning paths for TCM intelligent diagnosis.展开更多
Federated Learning(FL)protects data privacy through a distributed training mechanism,yet its decentralized nature also introduces new security vulnerabilities.Backdoor attacks inject malicious triggers into the global...Federated Learning(FL)protects data privacy through a distributed training mechanism,yet its decentralized nature also introduces new security vulnerabilities.Backdoor attacks inject malicious triggers into the global model through compromised updates,posing significant threats to model integrity and becoming a key focus in FL security.Existing backdoor attack methods typically embed triggers directly into original images and consider only data heterogeneity,resulting in limited stealth and adaptability.To address the heterogeneity of malicious client devices,this paper proposes a novel backdoor attack method named Capability-Adaptive Shadow Backdoor Attack(CASBA).By incorporating measurements of clients’computational and communication capabilities,CASBA employs a dynamic hierarchical attack strategy that adaptively aligns attack intensity with available resources.Furthermore,an improved deep convolutional generative adversarial network(DCGAN)is integrated into the attack pipeline to embed triggers without modifying original data,significantly enhancing stealthiness.Comparative experiments with Shadow Backdoor Attack(SBA)across multiple scenarios demonstrate that CASBA dynamically adjusts resource consumption based on device capabilities,reducing average memory usage per iteration by 5.8%.CASBA improves resource efficiency while keeping the drop in attack success rate within 3%.Additionally,the effectiveness of CASBA against three robust FL algorithms is also validated.展开更多
Accurate retrieval of atmospheric vertical profiles is critical for improving weather prediction and climate monitoring.However,the complexity of atmospheric processes in cloudy regions poses challenges compared to th...Accurate retrieval of atmospheric vertical profiles is critical for improving weather prediction and climate monitoring.However,the complexity of atmospheric processes in cloudy regions poses challenges compared to those of clear sky scenarios.This study presents a novel framework that integrates Bayesian optimization and machine learning approaches to retrieve atmospheric vertical profiles—including temperature,humidity,ozone concentration,cloud fraction,ice water content(IWC),and liquid water content(LWC)—from hyperspectral infrared observations.Specifically,a Bayesian method was used to refine ERA5 reanalysis data by minimizing brightness temperature(BT)discrepancies against FY-4B Geostationary Interferometric Infrared Sounder(GIIRS)observations,generating a high-quality profile database(~2.8 million profiles)across diverse weather systems.The optimized profiles improve radiative consistency,reducing BT biases from>40 K to<10 K in cloudy regions.To further overcome the limitations of the Bayesian method,we developed a Transformer-Resnet hybrid model(TERNet),which achieved superior performance with RMSE values of 1.61 K(temperature),5.77%(humidity),and 2.25×10^(–6)/6.09×10^(–6)kg kg^(–1)(IWC/LWC)across the entire vertical levels in all-sky conditions.The TERNet outperforms both ERA5 in cloud parameter retrieval and the GIIRS L2 product in thermodynamic profiling.Independent verification with radiosonde and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations(CALIPSO)datasets confirms the framework's reliability across various meteorological regimes.This work demonstrates the capability of combining physics-informed Bayesian methods with data-driven machine learning to fully exploit hyperspectral IR data.展开更多
Continual learning fault diagnosis(CLFD)has gained growing interest in mechanical systems for its ability to accumulate and transfer knowledge in dynamic fault diagnosis scenarios.However,existing CLFD methods typical...Continual learning fault diagnosis(CLFD)has gained growing interest in mechanical systems for its ability to accumulate and transfer knowledge in dynamic fault diagnosis scenarios.However,existing CLFD methods typically assume balanced task distributions,neglecting the long-tailed nature of real-world fault occurrences,where certain faults dominate while others are rare.Due to the long-tailed distribution among different me-chanical conditions,excessive attention has been focused on the dominant type,leading to performance de-gradation in rarer types.In this paper,decoupling incremental classifier and representation learning(DICRL)is proposed to address the dual challenges of catastrophic forgetting introduced by incremental tasks and the bias in long-tailed CLFD(LT-CLFD).The core innovation lies in the structural decoupling of incremental classifier learning and representation learning.An instance-balanced sampling strategy is employed to learn more dis-criminative deep representations from the exemplars selected by the herding algorithm and new data.Then,the previous classifiers are frozen to prevent damage to representation learning during backward propagation.Cosine normalization classifier with learnable weight scaling is trained using a class-balanced sampling strategy to enhance classification accuracy.Experimental results demonstrate that DICRL outperforms existing continual learning methods across multiple benchmarks,demonstrating superior performance and robustness in both LT-CLFD and conventional CLFD.DICRL effectively tackles both catastrophic forgetting and long-tailed distribution in CLFD,enabling more reliable fault diagnosis in industrial applications.展开更多
Landslides pose a formidable natural hazard across the Qinghai-Tibet Plateau(QTP),endangering both ecosystems and human life.Identifying the driving factors behind landslides and accurately assessing susceptibility ar...Landslides pose a formidable natural hazard across the Qinghai-Tibet Plateau(QTP),endangering both ecosystems and human life.Identifying the driving factors behind landslides and accurately assessing susceptibility are key to mitigating disaster risk.This study integrated multi-source historical landslide data with 15 predictive factors and used several machine learning models—Random Forest(RF),Gradient Boosting Regression Trees(GBRT),Extreme Gradient Boosting(XGBoost),and Categorical Boosting(CatBoost)—to generate susceptibility maps.The Shapley additive explanation(SHAP)method was applied to quantify factor importance and explore their nonlinear effects.The results showed that:(1)CatBoost was the best-performing model(CA=0.938,AUC=0.980)in assessing landslide susceptibility,with altitude emerging as the most significant factor,followed by distance to roads and earthquake sites,precipitation,and slope;(2)the SHAP method revealed critical nonlinear thresholds,demonstrating that historical landslides were concentrated at mid-altitudes(1400-4000 m)and decreased markedly above 4000 m,with a parallel reduction in probability beyond 700 m from roads;and(3)landslide-prone areas,comprising 13%of the QTP,were concentrated in the southeastern and northeastern parts of the plateau.By integrating machine learning and SHAP analysis,this study revealed landslide hazard-prone areas and their driving factors,providing insights to support disaster management strategies and sustainable regional planning.展开更多
The generation of synthetic trajectories has become essential in various fields for analyzing complex movement patterns.However,the use of real-world trajectory data poses significant privacy risks,such as location re...The generation of synthetic trajectories has become essential in various fields for analyzing complex movement patterns.However,the use of real-world trajectory data poses significant privacy risks,such as location reidentification and correlation attacks.To address these challenges,privacy-preserving trajectory generation methods are critical for applications relying on sensitive location data.This paper introduces DPIL-Traj,an advanced framework designed to generate synthetic trajectories while achieving a superior balance between data utility and privacy preservation.Firstly,the framework incorporates Differential Privacy Clustering,which anonymizes trajectory data by applying differential privacy techniques that add noise,ensuring the protection of sensitive user information.Secondly,Imitation Learning is used to replicate decision-making behaviors observed in real-world trajectories.By learning from expert trajectories,this component generates synthetic data that closely mimics real-world decision-making processes while optimizing the quality of the generated trajectories.Finally,Markov-based Trajectory Generation is employed to capture and maintain the inherent temporal dynamics of movement patterns.Extensive experiments conducted on the GeoLife trajectory dataset show that DPIL-Traj improves utility performance by an average of 19.85%,and in terms of privacy performance by an average of 12.51%,compared to state-of-the-art approaches.Ablation studies further reveal that DP clustering effectively safeguards privacy,imitation learning enhances utility under noise,and the Markov module strengthens temporal coherence.展开更多
Lithium manganese silicate(Li-Mn-Si-O)cathodes are key components of lithium-ion batteries,and their physical and mechanical properties are strongly influenced by their underlying crystal structures.In this study,a ra...Lithium manganese silicate(Li-Mn-Si-O)cathodes are key components of lithium-ion batteries,and their physical and mechanical properties are strongly influenced by their underlying crystal structures.In this study,a range of machine learning(ML)algorithms were developed and compared to predict the crystal systems of Li-Mn-Si-O cathode materials using density functional theory(DFT)data obtained from the Materials Project database.The dataset comprised 211 compositions characterized by key descriptors,including formation energy,energy above the hull,bandgap,atomic site number,density,and unit cell volume.These features were utilized to classify the materials into monoclinic(0)and triclinic(1)crystal systems.A comprehensive comparison of various classification algorithms including Decision Tree,Random Forest,XGBoost,Support VectorMachine,k-Nearest Neighbor,Stochastic Gradient Descent,Gaussian Naive Bayes,Gaussian Process,and Artificial Neural Network(ANN)was conducted.Among these,the optimized ANN architecture(6–14-14-14-1)exhibited the highest predictive performance,achieving an accuracy of 95.3%,aMatthews correlation coefficient(MCC)of 0.894,and an F-score of 0.963,demonstrating excellent consistency with DFT-predicted crystal structures.Meanwhile,RandomForest and Gaussian Processmodels also exhibited reliable and consistent predictive capability,indicating their potential as complementary approaches,particularly when data are limited or computational efficiency is required.This comparative framework provides valuable insights into model selection for crystal system classification in complex cathode materials.展开更多
Modern intelligent systems,such as autonomous vehicles and face recognition,must continuously adapt to new scenarios while preserving their ability to handle previously encountered situations.However,when neural netwo...Modern intelligent systems,such as autonomous vehicles and face recognition,must continuously adapt to new scenarios while preserving their ability to handle previously encountered situations.However,when neural networks learn new classes sequentially,they suffer from catastrophic forgetting—the tendency to lose knowledge of earlier classes.This challenge,which lies at the core of class-incremental learning,severely limits the deployment of continual learning systems in real-world applications with streaming data.Existing approaches,including rehearsalbased methods and knowledge distillation techniques,have attempted to address this issue but often struggle to effectively preserve decision boundaries and discriminative features under limited memory constraints.To overcome these limitations,we propose a support vector-guided framework for class-incremental learning.The framework integrates an enhanced feature extractor with a Support Vector Machine classifier,which generates boundary-critical support vectors to guide both replay and distillation.Building on this architecture,we design a joint feature retention strategy that combines boundary proximity with feature diversity,and a Support Vector Distillation Loss that enforces dual alignment in decision and semantic spaces.In addition,triple attention modules are incorporated into the feature extractor to enhance representation power.Extensive experiments on CIFAR-100 and Tiny-ImageNet demonstrate effective improvements.On CIFAR-100 and Tiny-ImageNet with 5 tasks,our method achieves 71.68%and 58.61%average accuracy,outperforming strong baselines by 3.34%and 2.05%.These advantages are consistently observed across different task splits,highlighting the robustness and generalization of the proposed approach.Beyond benchmark evaluations,the framework also shows potential in few-shot and resource-constrained applications such as edge computing and mobile robotics.展开更多
Deep learning and fatigue life prediction remain focal research areas in rail vehicle engineering.This study addresses the vibration fatigue of wheelset lifting lug in Chengdu Metro Line 1 bogies,aiming to develop a f...Deep learning and fatigue life prediction remain focal research areas in rail vehicle engineering.This study addresses the vibration fatigue of wheelset lifting lug in Chengdu Metro Line 1 bogies,aiming to develop a fatigue life prediction method for critical bogie components using deep learning models and measured track load spectra.Extensive field tests on Chengdu Metro Line 1 were conducted to acquire acceleration and stress response data of the wheelset lifting lug,generating training samples for the neural network system.Component stress responses were calculated via time-domain track acceleration and validated against in-situ stress measurements.Results show that neural network-fitted dynamic stress values exhibit excellent consistency with measured data,with errors constrained within 5%.This study validates the proposed small-sample deep learning approach as an effective and accurate solution for fatigue life prediction of critical bogie components under operational load conditions.展开更多
Structural Health Monitoring(SHM)plays a critical role in ensuring the safety,integrity,longevity and economic efficiency of civil infrastructures.The field has undergone a profound transformation over the last few de...Structural Health Monitoring(SHM)plays a critical role in ensuring the safety,integrity,longevity and economic efficiency of civil infrastructures.The field has undergone a profound transformation over the last few decades,evolving from traditional methods—often reliant on visual inspections—to data-driven intelligent systems.This review paper analyzes this historical trajectory,beginning with the approaches that relied on modal parameters as primary damage indicators.The advent of advanced sensor technologies and increased computational power brings a significant change,making Machine Learning(ML)a viable and powerful tool for damage assessment.More recently,Deep Learning(DL)has emerged as a paradigm shift,allowing for more automated processing of large data sets(such as the structural vibration signals and other types of sensors)with excellent performance and accuracy,often surpassing previous methods.This paper systematically reviews these technological milestones—from traditional vibration-based methods to the current state-of-the-art in deep learning.Finally,it critically examines emerging trends—such as Digital Twins and Transformer-based architectures—and discusses future research directions that will shape the next generation of SHM systems for civil engineering.展开更多
Aircraft assembly is characterized by stringent precedence constraints,limited resource availability,spatial restrictions,and a high degree of manual intervention.These factors lead to considerable variability in oper...Aircraft assembly is characterized by stringent precedence constraints,limited resource availability,spatial restrictions,and a high degree of manual intervention.These factors lead to considerable variability in operator workloads and significantly increase the complexity of scheduling.To address this challenge,this study investigates the Aircraft Pulsating Assembly Line Scheduling Problem(APALSP)under skilled operator allocation,with the objective of minimizing assembly completion time.A mathematical model considering skilled operator allocation is developed,and a Q-Learning improved Particle Swarm Optimization algorithm(QLPSO)is proposed.In the algorithm design,a reverse scheduling strategy is adopted to effectively manage large-scale precedence constraints.Moreover,a reverse sequence encoding method is introduced to generate operation sequences,while a time decoding mechanism is employed to determine completion times.The problem is further reformulated as a Markov Decision Process(MDP)with explicitly defined state and action spaces.Within QLPSO,the Q-learning mechanism adaptively adjusts inertia weights and learning factors,thereby achieving a balance between exploration capability and convergence performance.To validate the effectiveness of the proposed approach,extensive computational experiments are conducted on benchmark instances of different scales,including small,medium,large,and ultra-large cases.The results demonstrate that QLPSO consistently delivers stable and high-quality solutions across all scenarios.In ultra-large-scale instances,it improves the best solution by 25.2%compared with the Genetic Algorithm(GA)and enhances the average solution by 16.9%over the Q-learning algorithm,showing clear advantages over the comparative methods.These findings not only confirm the effectiveness of the proposed algorithm but also provide valuable theoretical references and practical guidance for the intelligent scheduling optimization of aircraft pulsating assembly lines.展开更多
Artificial Intelligence(AI)is changing healthcare by helping with diagnosis.However,for doctors to trust AI tools,they need to be both accurate and easy to understand.In this study,we created a new machine learning sy...Artificial Intelligence(AI)is changing healthcare by helping with diagnosis.However,for doctors to trust AI tools,they need to be both accurate and easy to understand.In this study,we created a new machine learning system for the early detection of Autism Spectrum Disorder(ASD)in children.Our main goal was to build a model that is not only good at predicting ASD but also clear in its reasoning.For this,we combined several different models,including Random Forest,XGBoost,and Neural Networks,into a single,more powerful framework.We used two different types of datasets:(i)a standard behavioral dataset and(ii)a more complex multimodal dataset with images,audio,and physiological information.The datasets were carefully preprocessed for missing values,redundant features,and dataset imbalance to ensure fair learning.The results outperformed the state-of-the-art with a Regularized Neural Network,achieving 97.6%accuracy on behavioral data.Whereas,on the multimodal data,the accuracy is 98.2%.Other models also did well with accuracies consistently above 96%.We also used SHAP and LIME on a behavioral dataset for models’explainability.展开更多
This study examines how foreign language education in the artificial intelligence(AI)era could assist the cultivation of national consciousness through a technology-enhanced pedagogy of film appreciation.Using The Wil...This study examines how foreign language education in the artificial intelligence(AI)era could assist the cultivation of national consciousness through a technology-enhanced pedagogy of film appreciation.Using The Wild Robot as a case study,we argue that cinematic narratives serve as cultural mirrors,offering immersive,reflective,and affective sites for intercultural learning.We propose a three-layered pedagogical framework-progressing from semiotic decoding,through narrative and value comparison,to creative identity construction-that integrates intelligent tools to develop both communicative competence and an agentive sense of belonging.The approach exemplifies a humanistic turn in language teaching,aiming to form“rooted global communicators”who can engage in cross-civilization dialogue with cultural confidence and critical awareness.展开更多
Professional learning communities(PLCs)offer essential contextual support for the development of foreign language teachers in higher education.The book Building Professional Learning Communities of Foreign Language Te...Professional learning communities(PLCs)offer essential contextual support for the development of foreign language teachers in higher education.The book Building Professional Learning Communities of Foreign Language Teachers in Higher Education,co-authored by Wen et al.(2021)systematically examines the necessity,practical measures,theoretical construction,and outcomes associated with building PLCs for university foreign language teachers.Notably,it introduces a research methodology rooted in local teacher education practices with Chinese characteristics—the dialectical research paradigm(DRP).This review introduces the content of the book,evaluates its contributions to teacher development,and explores its implications for future practice and research.展开更多
Lithology identificationwhile drilling technology can obtain rock information in real-time.However,traditional lithology identificationmodels often face limitations in feature extraction and adaptability to complex ge...Lithology identificationwhile drilling technology can obtain rock information in real-time.However,traditional lithology identificationmodels often face limitations in feature extraction and adaptability to complex geological conditions,limiting their accuracy in challenging environments.To address these challenges,a deep learning model for lithology identificationwhile drilling is proposed.The proposed model introduces a dual attention mechanism in the long short-term memory(LSTM)network,effectively enhancing the ability to capture spatial and channel dimension information.Subsequently,the crayfishoptimization algorithm(COA)is applied to optimize the model network structure,thereby enhancing its lithology identificationcapability.Laboratory test results demonstrate that the proposed model achieves 97.15%accuracy on the testing set,significantlyoutperforming the traditional support vector machine(SVM)method(81.77%).Field tests under actual drilling conditions demonstrate an average accuracy of 91.96%for the proposed model,representing a 14.31%improvement over the LSTM model alone.The proposed model demonstrates robust adaptability and generalization ability across diverse operational scenarios.This research offers reliable technical support for lithology identification while drilling.展开更多
At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown ...At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments,this paper proposes an Attention-Enhanced Dueling Deep Q-Network(ADDueling DQN),which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework.A multi-objective reward function,centered on energy efficiency,is designed to comprehensively consider path length,terrain slope,motion smoothness,and obstacle avoidance,enabling optimal low-energy trajectory generation in 3D space from the source.The incorporation of a multihead attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths.Additionally,the prioritized experience replay mechanism accelerates learning from key decision-making experiences,suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly.The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios.Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments.Moreover,the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms,highlighting its global optimization capability under energy-aware objectives in complex terrains.This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.展开更多
With an increase in internet-connected devices and a dependency on online services,the threat of Distributed Denial of Service(DDoS)attacks has become a significant concern in cybersecurity.The proposed system follows...With an increase in internet-connected devices and a dependency on online services,the threat of Distributed Denial of Service(DDoS)attacks has become a significant concern in cybersecurity.The proposed system follows a multi-step process,beginning with the collection of datasets from different edge devices and network nodes.To verify its effectiveness,experiments were conducted using the CICDoS2017,NSL-KDD,and CICIDS benchmark datasets alongside other existing models.Recursive feature elimination(RFE)with random forest is used to select features from the CICDDoS2019 dataset,on which a BiLSTM model is trained on local nodes.Local models are trained until convergence or stability criteria are met while simultaneously sharing the updates globally for collaborative learning.A centralised server evaluates real-time traffic using the global BiLSTM model,which triggers alerts for potential DDoS attacks.Furthermore,blockchain technology is employed to secure model updates and to provide an immutable audit trail,thereby ensuring trust and accountability among network nodes.This research introduces a novel decentralized method called Federated Random Forest Bidirectional Long Short-Term Memory(FRF-BiLSTM)for detecting DDoS attacks,utilizing the advanced Bidirectional Long Short-Term Memory Networks(BiLSTMs)to analyze sequences in both forward and backward directions.The outcome shows the proposed model achieves a mean accuracy of 97.1%with an average training delay of 88.7 s and testing delay of 21.4 s.The model demonstrates scalability and the best detection performance in large-scale attack scenarios.展开更多
Landslide susceptibility mapping(LSM)is an essential tool for mitigating the escalating global risk of landslides.However,challenges such as the heterogeneity of different landslide triggers,extensive engineering acti...Landslide susceptibility mapping(LSM)is an essential tool for mitigating the escalating global risk of landslides.However,challenges such as the heterogeneity of different landslide triggers,extensive engineering activities exacerbated reactivation,and the interpretability of data-driven models have hindered the practical application of LSM.This work proposes a novel framework for enhancing LSM considering different triggers for accumulation and rock landslides,leveraging interpretable machine learning and Multi-temporal Interferometric Synthetic Aperture Radar(MT-InSAR)technology.Initially,a refined fieldinvestigation was conducted to delineate the accumulation and rock area according to landslide types,leading to the identificationof relevant contributing factors.Deformation along the slope was then combined with time-series analysis to derive a landslide activity level(AL)index to recognize the likelihood of reactivation or dormancy.The SHapley Additive exPlanation(SHAP)technique facilitated the interpretation of factors and the identificationof determinants in high susceptibility areas.The results indicate that random forest(RF)outperformed other models in both accumulation and rock areas.Key factors including thickness and weak intercalation were identifiedfor accumulation and rock landslides.The introduction of AL substantially enhanced the predictive capability of the LSM and outperformed models that neglect movement trends or deformation rates with an average ratio of 81.23%in high susceptibility zones.Besides,the fieldvalidation confirmedthat 83.8%of newly identifiedlandslides were correctly upgraded.Given its efficiencyand operational simplicity,the proposed hybrid model opens new avenues for the feasibility of enhancement in LSM at urban settlements worldwide.展开更多
This study developed a modeling methodology for statistical optimization-based geologic hazard susceptibility assessment,aiming to enhance the comprehensive performance and classification accuracy of the assessment mo...This study developed a modeling methodology for statistical optimization-based geologic hazard susceptibility assessment,aiming to enhance the comprehensive performance and classification accuracy of the assessment models.First,the cumulative probability method revealed that a low probability(15%)of geologic hazards between any two geologic hazard points occurred outside a buffer zone with a radius of 2297 m(i.e.,the distance threshold).The training dataset was established,consisting of negative samples(non-hazard points)randomly generated based on the distance threshold,positive samples(i.e.,historical hazards),and 13 conditioning factors.Then,models were built using five machine learning algorithms,namely random forest(RF),gradient boosting decision tree(GBDT),naive Bayes(NB),logistic regression(LR),and support vector machine(SVM).The comprehensive performance of the models was assessed using the area under the receiver operating characteristic curve(AUC)and overall accuracy(OA)as indicators,revealing that RF exhibited the best performance,with OA and AUC values of 2.7127 and 0.981,respectively.Furthermore,the machine learning models constructed by considering the distance threshold outperformed those built using the unoptimized dataset.The characteristic factors were ranked using the mutual information method,with their scores decreasing in the order of rainfall(0.1616),altitude(0.06),normalized difference vegetation index(NDVI;0.04),and distance from roads(0.03).Finally,the geologic hazard susceptibility classification was assessed using the natural breaks method combined with a clustering algorithm.The results indicate that the clustering algorithm exhibited higher classification accuracy than the natural breaks method.The findings of this study demonstrate that the proposed model optimization scheme can provide a scientific basis for the prevention and control of geologic hazards.展开更多
文摘Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.
基金Sichuan TCM Culture Coordinated Development Research Center Project(2023XT131)National Key Science and Technology Project of China(2023ZD0509405)National Natural Science Foundation of China(82174236).
文摘Objective To address the dual challenges of long-tail distribution and feature sparsity in traditional Chinese medicine(TCM)syndrome differentiation within real clinical settings,we propose a data-efficient learning framework enhanced by knowledge graphs.Methods We developed Agent-GNN,a three-stage decoupled learning framework,and validated it on the Traditional Chinese Medicine Syndrome Diagnosis(TCM-SD)dataset containing 54152 clinical records across 148 syndrome categories.First,we constructed a comprehensive medical knowledge graph encoding the complete TCM reasoning system.Second,we proposed a Functional Patient Profiling(FPP)method that utilizes large language models(LLMs)combined with Graph Retrieval-Augmented Generation(RAG)to extract structured symptom-etiology-pathogenesis subgraphs from medical records.Third,we employed heterogeneous graph neural networks to learn structured combination patterns explicitly.We compared our method against multiple baselines including BERT,ZY-BERT,ZY-BERT+Know,GAT,and GPT-4 Few-shot,using macro-F1 score as the primary evaluation metric.Additionally,ablation experiments were conducted to validate the contribution of each key component to model performance.Results Agent-GNN achieved an overall macro-F1 score of 72.4%,representing an 8.7 percentage points improvement over ZY-BERT+Know(63.7%),the strongest baseline among traditional methods.For long-tail syndromes with fewer than 10 samples,Agent-GNN reached a macro-F1 score of 58.6%,compared with 39.3%for ZY-BERT+Know and 41.2%for GPT-4 Few-shot,representing relative improvements of 49.2%and 42.2%,respectively.Ablation experiments confirmed that the explicit modeling of etiology-pathogenesis nodes contributed 12.4 percentage points to this enhanced long-tail syndrome performance.Conclusion This study proposes Agent-GNN,a knowledge graph-enhanced framework that effectively addresses the long-tail distribution challenge in TCM syndrome differentiation.By explicitly modeling manifestation-mechanism-essence patterns through structured knowledge graphs,our approach achieves superior performance in data-scarce scenarios while providing interpretable reasoning paths for TCM intelligent diagnosis.
基金supported by the National Natural Science Foundation of China(Grant No.62172123)the Key Research and Development Program of Heilongjiang Province,China(GrantNo.2022ZX01A36).
文摘Federated Learning(FL)protects data privacy through a distributed training mechanism,yet its decentralized nature also introduces new security vulnerabilities.Backdoor attacks inject malicious triggers into the global model through compromised updates,posing significant threats to model integrity and becoming a key focus in FL security.Existing backdoor attack methods typically embed triggers directly into original images and consider only data heterogeneity,resulting in limited stealth and adaptability.To address the heterogeneity of malicious client devices,this paper proposes a novel backdoor attack method named Capability-Adaptive Shadow Backdoor Attack(CASBA).By incorporating measurements of clients’computational and communication capabilities,CASBA employs a dynamic hierarchical attack strategy that adaptively aligns attack intensity with available resources.Furthermore,an improved deep convolutional generative adversarial network(DCGAN)is integrated into the attack pipeline to embed triggers without modifying original data,significantly enhancing stealthiness.Comparative experiments with Shadow Backdoor Attack(SBA)across multiple scenarios demonstrate that CASBA dynamically adjusts resource consumption based on device capabilities,reducing average memory usage per iteration by 5.8%.CASBA improves resource efficiency while keeping the drop in attack success rate within 3%.Additionally,the effectiveness of CASBA against three robust FL algorithms is also validated.
基金supported by the National Natural Science Foundation of China under Grant U2442219Fengyun Satellite Application Pioneer Program(2023)Special Initiative on Numerical Weather Prediction(NWP)Applications,the Civil Aerospace Technology Pre-Research Project(D040405)the Joint Funds of the Zhejiang Provincial Natural Science Foundation of China under Grant No.LZJMZ23D050003。
文摘Accurate retrieval of atmospheric vertical profiles is critical for improving weather prediction and climate monitoring.However,the complexity of atmospheric processes in cloudy regions poses challenges compared to those of clear sky scenarios.This study presents a novel framework that integrates Bayesian optimization and machine learning approaches to retrieve atmospheric vertical profiles—including temperature,humidity,ozone concentration,cloud fraction,ice water content(IWC),and liquid water content(LWC)—from hyperspectral infrared observations.Specifically,a Bayesian method was used to refine ERA5 reanalysis data by minimizing brightness temperature(BT)discrepancies against FY-4B Geostationary Interferometric Infrared Sounder(GIIRS)observations,generating a high-quality profile database(~2.8 million profiles)across diverse weather systems.The optimized profiles improve radiative consistency,reducing BT biases from>40 K to<10 K in cloudy regions.To further overcome the limitations of the Bayesian method,we developed a Transformer-Resnet hybrid model(TERNet),which achieved superior performance with RMSE values of 1.61 K(temperature),5.77%(humidity),and 2.25×10^(–6)/6.09×10^(–6)kg kg^(–1)(IWC/LWC)across the entire vertical levels in all-sky conditions.The TERNet outperforms both ERA5 in cloud parameter retrieval and the GIIRS L2 product in thermodynamic profiling.Independent verification with radiosonde and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations(CALIPSO)datasets confirms the framework's reliability across various meteorological regimes.This work demonstrates the capability of combining physics-informed Bayesian methods with data-driven machine learning to fully exploit hyperspectral IR data.
基金Supported by National Natural Science Foundation of China(Grant No.52272440)Suzhou Science Foundation(Grant Nos.SYG202323,ZXL2022027).
文摘Continual learning fault diagnosis(CLFD)has gained growing interest in mechanical systems for its ability to accumulate and transfer knowledge in dynamic fault diagnosis scenarios.However,existing CLFD methods typically assume balanced task distributions,neglecting the long-tailed nature of real-world fault occurrences,where certain faults dominate while others are rare.Due to the long-tailed distribution among different me-chanical conditions,excessive attention has been focused on the dominant type,leading to performance de-gradation in rarer types.In this paper,decoupling incremental classifier and representation learning(DICRL)is proposed to address the dual challenges of catastrophic forgetting introduced by incremental tasks and the bias in long-tailed CLFD(LT-CLFD).The core innovation lies in the structural decoupling of incremental classifier learning and representation learning.An instance-balanced sampling strategy is employed to learn more dis-criminative deep representations from the exemplars selected by the herding algorithm and new data.Then,the previous classifiers are frozen to prevent damage to representation learning during backward propagation.Cosine normalization classifier with learnable weight scaling is trained using a class-balanced sampling strategy to enhance classification accuracy.Experimental results demonstrate that DICRL outperforms existing continual learning methods across multiple benchmarks,demonstrating superior performance and robustness in both LT-CLFD and conventional CLFD.DICRL effectively tackles both catastrophic forgetting and long-tailed distribution in CLFD,enabling more reliable fault diagnosis in industrial applications.
基金The National Key Research and Development Program of China,No.2023YFC3206601。
文摘Landslides pose a formidable natural hazard across the Qinghai-Tibet Plateau(QTP),endangering both ecosystems and human life.Identifying the driving factors behind landslides and accurately assessing susceptibility are key to mitigating disaster risk.This study integrated multi-source historical landslide data with 15 predictive factors and used several machine learning models—Random Forest(RF),Gradient Boosting Regression Trees(GBRT),Extreme Gradient Boosting(XGBoost),and Categorical Boosting(CatBoost)—to generate susceptibility maps.The Shapley additive explanation(SHAP)method was applied to quantify factor importance and explore their nonlinear effects.The results showed that:(1)CatBoost was the best-performing model(CA=0.938,AUC=0.980)in assessing landslide susceptibility,with altitude emerging as the most significant factor,followed by distance to roads and earthquake sites,precipitation,and slope;(2)the SHAP method revealed critical nonlinear thresholds,demonstrating that historical landslides were concentrated at mid-altitudes(1400-4000 m)and decreased markedly above 4000 m,with a parallel reduction in probability beyond 700 m from roads;and(3)landslide-prone areas,comprising 13%of the QTP,were concentrated in the southeastern and northeastern parts of the plateau.By integrating machine learning and SHAP analysis,this study revealed landslide hazard-prone areas and their driving factors,providing insights to support disaster management strategies and sustainable regional planning.
基金supported by the Natural Science Foundation of Fujian Province of China(2025J01380)National Natural Science Foundation of China(No.62471139)+3 种基金the Major Health Research Project of Fujian Province(2021ZD01001)Fujian Provincial Units Special Funds for Education and Research(2022639)Fujian University of Technology Research Start-up Fund(GY-S24002)Fujian Research and Training Grants for Young and Middle-aged Leaders in Healthcare(GY-H-24179).
文摘The generation of synthetic trajectories has become essential in various fields for analyzing complex movement patterns.However,the use of real-world trajectory data poses significant privacy risks,such as location reidentification and correlation attacks.To address these challenges,privacy-preserving trajectory generation methods are critical for applications relying on sensitive location data.This paper introduces DPIL-Traj,an advanced framework designed to generate synthetic trajectories while achieving a superior balance between data utility and privacy preservation.Firstly,the framework incorporates Differential Privacy Clustering,which anonymizes trajectory data by applying differential privacy techniques that add noise,ensuring the protection of sensitive user information.Secondly,Imitation Learning is used to replicate decision-making behaviors observed in real-world trajectories.By learning from expert trajectories,this component generates synthetic data that closely mimics real-world decision-making processes while optimizing the quality of the generated trajectories.Finally,Markov-based Trajectory Generation is employed to capture and maintain the inherent temporal dynamics of movement patterns.Extensive experiments conducted on the GeoLife trajectory dataset show that DPIL-Traj improves utility performance by an average of 19.85%,and in terms of privacy performance by an average of 12.51%,compared to state-of-the-art approaches.Ablation studies further reveal that DP clustering effectively safeguards privacy,imitation learning enhances utility under noise,and the Markov module strengthens temporal coherence.
基金supported by the Learning&Academic Research Institution for Master’s,PhD students,and Postdocs LAMP Program of the National Research Foundation of Korea(NRF)grant funded by the Ministry of Education(No.RS-2023-00301974)This work was also supported by the Glocal University 30 Project fund of Gyeongsang National University in 2025.
文摘Lithium manganese silicate(Li-Mn-Si-O)cathodes are key components of lithium-ion batteries,and their physical and mechanical properties are strongly influenced by their underlying crystal structures.In this study,a range of machine learning(ML)algorithms were developed and compared to predict the crystal systems of Li-Mn-Si-O cathode materials using density functional theory(DFT)data obtained from the Materials Project database.The dataset comprised 211 compositions characterized by key descriptors,including formation energy,energy above the hull,bandgap,atomic site number,density,and unit cell volume.These features were utilized to classify the materials into monoclinic(0)and triclinic(1)crystal systems.A comprehensive comparison of various classification algorithms including Decision Tree,Random Forest,XGBoost,Support VectorMachine,k-Nearest Neighbor,Stochastic Gradient Descent,Gaussian Naive Bayes,Gaussian Process,and Artificial Neural Network(ANN)was conducted.Among these,the optimized ANN architecture(6–14-14-14-1)exhibited the highest predictive performance,achieving an accuracy of 95.3%,aMatthews correlation coefficient(MCC)of 0.894,and an F-score of 0.963,demonstrating excellent consistency with DFT-predicted crystal structures.Meanwhile,RandomForest and Gaussian Processmodels also exhibited reliable and consistent predictive capability,indicating their potential as complementary approaches,particularly when data are limited or computational efficiency is required.This comparative framework provides valuable insights into model selection for crystal system classification in complex cathode materials.
基金supported by the Gansu Provincial Natural Science Foundation(grant number 25JRRA074)the Gansu Provincial Key R&D Science and Technology Program(grant number 24YFGA060)the National Natural Science Foundation of China(grant number 62161019).
文摘Modern intelligent systems,such as autonomous vehicles and face recognition,must continuously adapt to new scenarios while preserving their ability to handle previously encountered situations.However,when neural networks learn new classes sequentially,they suffer from catastrophic forgetting—the tendency to lose knowledge of earlier classes.This challenge,which lies at the core of class-incremental learning,severely limits the deployment of continual learning systems in real-world applications with streaming data.Existing approaches,including rehearsalbased methods and knowledge distillation techniques,have attempted to address this issue but often struggle to effectively preserve decision boundaries and discriminative features under limited memory constraints.To overcome these limitations,we propose a support vector-guided framework for class-incremental learning.The framework integrates an enhanced feature extractor with a Support Vector Machine classifier,which generates boundary-critical support vectors to guide both replay and distillation.Building on this architecture,we design a joint feature retention strategy that combines boundary proximity with feature diversity,and a Support Vector Distillation Loss that enforces dual alignment in decision and semantic spaces.In addition,triple attention modules are incorporated into the feature extractor to enhance representation power.Extensive experiments on CIFAR-100 and Tiny-ImageNet demonstrate effective improvements.On CIFAR-100 and Tiny-ImageNet with 5 tasks,our method achieves 71.68%and 58.61%average accuracy,outperforming strong baselines by 3.34%and 2.05%.These advantages are consistently observed across different task splits,highlighting the robustness and generalization of the proposed approach.Beyond benchmark evaluations,the framework also shows potential in few-shot and resource-constrained applications such as edge computing and mobile robotics.
基金supported by the CRRC Original Technology TenYear Cultivation Program(Grant No.2022CYY007)。
文摘Deep learning and fatigue life prediction remain focal research areas in rail vehicle engineering.This study addresses the vibration fatigue of wheelset lifting lug in Chengdu Metro Line 1 bogies,aiming to develop a fatigue life prediction method for critical bogie components using deep learning models and measured track load spectra.Extensive field tests on Chengdu Metro Line 1 were conducted to acquire acceleration and stress response data of the wheelset lifting lug,generating training samples for the neural network system.Component stress responses were calculated via time-domain track acceleration and validated against in-situ stress measurements.Results show that neural network-fitted dynamic stress values exhibit excellent consistency with measured data,with errors constrained within 5%.This study validates the proposed small-sample deep learning approach as an effective and accurate solution for fatigue life prediction of critical bogie components under operational load conditions.
基金The authors would like to thank CNPq(Conselho Nacional de Desenvolvimento Científico e Tecnológico)—grants 407256/2022-9,303550/2025-2,402533/2023-2 and 303982/2022-5FAPEMIG(Fundação de AmparoàPesquisa do Estado de Minas Gerais)—grants APQ-00032-24 and APD-01113-25 for their financial support.
文摘Structural Health Monitoring(SHM)plays a critical role in ensuring the safety,integrity,longevity and economic efficiency of civil infrastructures.The field has undergone a profound transformation over the last few decades,evolving from traditional methods—often reliant on visual inspections—to data-driven intelligent systems.This review paper analyzes this historical trajectory,beginning with the approaches that relied on modal parameters as primary damage indicators.The advent of advanced sensor technologies and increased computational power brings a significant change,making Machine Learning(ML)a viable and powerful tool for damage assessment.More recently,Deep Learning(DL)has emerged as a paradigm shift,allowing for more automated processing of large data sets(such as the structural vibration signals and other types of sensors)with excellent performance and accuracy,often surpassing previous methods.This paper systematically reviews these technological milestones—from traditional vibration-based methods to the current state-of-the-art in deep learning.Finally,it critically examines emerging trends—such as Digital Twins and Transformer-based architectures—and discusses future research directions that will shape the next generation of SHM systems for civil engineering.
基金supported by the National Natural Science Foundation of China(Grant No.52475543)Natural Science Foundation of Henan(Grant No.252300421101)+1 种基金Henan Province University Science and Technology Innovation Talent Support Plan(Grant No.24HASTIT048)Science and Technology Innovation Team Project of Zhengzhou University of Light Industry(Grant No.23XNKJTD0101).
文摘Aircraft assembly is characterized by stringent precedence constraints,limited resource availability,spatial restrictions,and a high degree of manual intervention.These factors lead to considerable variability in operator workloads and significantly increase the complexity of scheduling.To address this challenge,this study investigates the Aircraft Pulsating Assembly Line Scheduling Problem(APALSP)under skilled operator allocation,with the objective of minimizing assembly completion time.A mathematical model considering skilled operator allocation is developed,and a Q-Learning improved Particle Swarm Optimization algorithm(QLPSO)is proposed.In the algorithm design,a reverse scheduling strategy is adopted to effectively manage large-scale precedence constraints.Moreover,a reverse sequence encoding method is introduced to generate operation sequences,while a time decoding mechanism is employed to determine completion times.The problem is further reformulated as a Markov Decision Process(MDP)with explicitly defined state and action spaces.Within QLPSO,the Q-learning mechanism adaptively adjusts inertia weights and learning factors,thereby achieving a balance between exploration capability and convergence performance.To validate the effectiveness of the proposed approach,extensive computational experiments are conducted on benchmark instances of different scales,including small,medium,large,and ultra-large cases.The results demonstrate that QLPSO consistently delivers stable and high-quality solutions across all scenarios.In ultra-large-scale instances,it improves the best solution by 25.2%compared with the Genetic Algorithm(GA)and enhances the average solution by 16.9%over the Q-learning algorithm,showing clear advantages over the comparative methods.These findings not only confirm the effectiveness of the proposed algorithm but also provide valuable theoretical references and practical guidance for the intelligent scheduling optimization of aircraft pulsating assembly lines.
基金the King Salman center for Disability Research for funding this work through Research Group No.KSRG-2024-050.
文摘Artificial Intelligence(AI)is changing healthcare by helping with diagnosis.However,for doctors to trust AI tools,they need to be both accurate and easy to understand.In this study,we created a new machine learning system for the early detection of Autism Spectrum Disorder(ASD)in children.Our main goal was to build a model that is not only good at predicting ASD but also clear in its reasoning.For this,we combined several different models,including Random Forest,XGBoost,and Neural Networks,into a single,more powerful framework.We used two different types of datasets:(i)a standard behavioral dataset and(ii)a more complex multimodal dataset with images,audio,and physiological information.The datasets were carefully preprocessed for missing values,redundant features,and dataset imbalance to ensure fair learning.The results outperformed the state-of-the-art with a Regularized Neural Network,achieving 97.6%accuracy on behavioral data.Whereas,on the multimodal data,the accuracy is 98.2%.Other models also did well with accuracies consistently above 96%.We also used SHAP and LIME on a behavioral dataset for models’explainability.
基金supported by the project:Hunan Provincial Educational Science Research Project“Research on Cultivating National Consciousness in College Foreign Language Courses(XJT23CGD001)”.
文摘This study examines how foreign language education in the artificial intelligence(AI)era could assist the cultivation of national consciousness through a technology-enhanced pedagogy of film appreciation.Using The Wild Robot as a case study,we argue that cinematic narratives serve as cultural mirrors,offering immersive,reflective,and affective sites for intercultural learning.We propose a three-layered pedagogical framework-progressing from semiotic decoding,through narrative and value comparison,to creative identity construction-that integrates intelligent tools to develop both communicative competence and an agentive sense of belonging.The approach exemplifies a humanistic turn in language teaching,aiming to form“rooted global communicators”who can engage in cross-civilization dialogue with cultural confidence and critical awareness.
文摘Professional learning communities(PLCs)offer essential contextual support for the development of foreign language teachers in higher education.The book Building Professional Learning Communities of Foreign Language Teachers in Higher Education,co-authored by Wen et al.(2021)systematically examines the necessity,practical measures,theoretical construction,and outcomes associated with building PLCs for university foreign language teachers.Notably,it introduces a research methodology rooted in local teacher education practices with Chinese characteristics—the dialectical research paradigm(DRP).This review introduces the content of the book,evaluates its contributions to teacher development,and explores its implications for future practice and research.
基金supported by the National Key Research and Development Program for Young Scientists,Chin(Grant No.2021YFC2900400)the Sichuan-Chongqing Science and Technology Innovation Cooperation Program Project,China(Grant No.2024TIAD-CYKJCXX0269)the National Natural Science Foundation of China,China(Grant No.52304123).
文摘Lithology identificationwhile drilling technology can obtain rock information in real-time.However,traditional lithology identificationmodels often face limitations in feature extraction and adaptability to complex geological conditions,limiting their accuracy in challenging environments.To address these challenges,a deep learning model for lithology identificationwhile drilling is proposed.The proposed model introduces a dual attention mechanism in the long short-term memory(LSTM)network,effectively enhancing the ability to capture spatial and channel dimension information.Subsequently,the crayfishoptimization algorithm(COA)is applied to optimize the model network structure,thereby enhancing its lithology identificationcapability.Laboratory test results demonstrate that the proposed model achieves 97.15%accuracy on the testing set,significantlyoutperforming the traditional support vector machine(SVM)method(81.77%).Field tests under actual drilling conditions demonstrate an average accuracy of 91.96%for the proposed model,representing a 14.31%improvement over the LSTM model alone.The proposed model demonstrates robust adaptability and generalization ability across diverse operational scenarios.This research offers reliable technical support for lithology identification while drilling.
文摘At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments,this paper proposes an Attention-Enhanced Dueling Deep Q-Network(ADDueling DQN),which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework.A multi-objective reward function,centered on energy efficiency,is designed to comprehensively consider path length,terrain slope,motion smoothness,and obstacle avoidance,enabling optimal low-energy trajectory generation in 3D space from the source.The incorporation of a multihead attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths.Additionally,the prioritized experience replay mechanism accelerates learning from key decision-making experiences,suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly.The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios.Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments.Moreover,the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms,highlighting its global optimization capability under energy-aware objectives in complex terrains.This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.
基金supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea(NRF-2025S1A5A2A01005171)by the BK21 programat Chungbuk National University(2025).
文摘With an increase in internet-connected devices and a dependency on online services,the threat of Distributed Denial of Service(DDoS)attacks has become a significant concern in cybersecurity.The proposed system follows a multi-step process,beginning with the collection of datasets from different edge devices and network nodes.To verify its effectiveness,experiments were conducted using the CICDoS2017,NSL-KDD,and CICIDS benchmark datasets alongside other existing models.Recursive feature elimination(RFE)with random forest is used to select features from the CICDDoS2019 dataset,on which a BiLSTM model is trained on local nodes.Local models are trained until convergence or stability criteria are met while simultaneously sharing the updates globally for collaborative learning.A centralised server evaluates real-time traffic using the global BiLSTM model,which triggers alerts for potential DDoS attacks.Furthermore,blockchain technology is employed to secure model updates and to provide an immutable audit trail,thereby ensuring trust and accountability among network nodes.This research introduces a novel decentralized method called Federated Random Forest Bidirectional Long Short-Term Memory(FRF-BiLSTM)for detecting DDoS attacks,utilizing the advanced Bidirectional Long Short-Term Memory Networks(BiLSTMs)to analyze sequences in both forward and backward directions.The outcome shows the proposed model achieves a mean accuracy of 97.1%with an average training delay of 88.7 s and testing delay of 21.4 s.The model demonstrates scalability and the best detection performance in large-scale attack scenarios.
基金supported by the National Key R&D Program of China(Grant No.2023YFC3007201)the National Natural Science Foundation of China(Grant No.42377161)the Opening Fund of Key Laboratory of Geological Survey and Evaluation of Ministry of Education(Grant No.GLAB 2024ZR03).
文摘Landslide susceptibility mapping(LSM)is an essential tool for mitigating the escalating global risk of landslides.However,challenges such as the heterogeneity of different landslide triggers,extensive engineering activities exacerbated reactivation,and the interpretability of data-driven models have hindered the practical application of LSM.This work proposes a novel framework for enhancing LSM considering different triggers for accumulation and rock landslides,leveraging interpretable machine learning and Multi-temporal Interferometric Synthetic Aperture Radar(MT-InSAR)technology.Initially,a refined fieldinvestigation was conducted to delineate the accumulation and rock area according to landslide types,leading to the identificationof relevant contributing factors.Deformation along the slope was then combined with time-series analysis to derive a landslide activity level(AL)index to recognize the likelihood of reactivation or dormancy.The SHapley Additive exPlanation(SHAP)technique facilitated the interpretation of factors and the identificationof determinants in high susceptibility areas.The results indicate that random forest(RF)outperformed other models in both accumulation and rock areas.Key factors including thickness and weak intercalation were identifiedfor accumulation and rock landslides.The introduction of AL substantially enhanced the predictive capability of the LSM and outperformed models that neglect movement trends or deformation rates with an average ratio of 81.23%in high susceptibility zones.Besides,the fieldvalidation confirmedthat 83.8%of newly identifiedlandslides were correctly upgraded.Given its efficiencyand operational simplicity,the proposed hybrid model opens new avenues for the feasibility of enhancement in LSM at urban settlements worldwide.
基金supported by a project entitled Loess Plateau Region-Watershed-Slope Geological Hazard Multi-Scale Collaborative Intelligent Early Warning System of the National Key R&D Program of China(2022YFC3003404)a project of the Shaanxi Youth Science and Technology Star(2021KJXX-87)public welfare geological survey projects of Shaanxi Institute of Geologic Survey(20180301,201918,202103,and 202413)。
文摘This study developed a modeling methodology for statistical optimization-based geologic hazard susceptibility assessment,aiming to enhance the comprehensive performance and classification accuracy of the assessment models.First,the cumulative probability method revealed that a low probability(15%)of geologic hazards between any two geologic hazard points occurred outside a buffer zone with a radius of 2297 m(i.e.,the distance threshold).The training dataset was established,consisting of negative samples(non-hazard points)randomly generated based on the distance threshold,positive samples(i.e.,historical hazards),and 13 conditioning factors.Then,models were built using five machine learning algorithms,namely random forest(RF),gradient boosting decision tree(GBDT),naive Bayes(NB),logistic regression(LR),and support vector machine(SVM).The comprehensive performance of the models was assessed using the area under the receiver operating characteristic curve(AUC)and overall accuracy(OA)as indicators,revealing that RF exhibited the best performance,with OA and AUC values of 2.7127 and 0.981,respectively.Furthermore,the machine learning models constructed by considering the distance threshold outperformed those built using the unoptimized dataset.The characteristic factors were ranked using the mutual information method,with their scores decreasing in the order of rainfall(0.1616),altitude(0.06),normalized difference vegetation index(NDVI;0.04),and distance from roads(0.03).Finally,the geologic hazard susceptibility classification was assessed using the natural breaks method combined with a clustering algorithm.The results indicate that the clustering algorithm exhibited higher classification accuracy than the natural breaks method.The findings of this study demonstrate that the proposed model optimization scheme can provide a scientific basis for the prevention and control of geologic hazards.
