Federated learning is a distributed framework that trains a centralised model using data from multiple clients without transferring that data to a central server.Despite rapid progress,federated learning still faces s...Federated learning is a distributed framework that trains a centralised model using data from multiple clients without transferring that data to a central server.Despite rapid progress,federated learning still faces several unsolved challenges.Specifically,communication costs and system heterogeneity,such as nonidentical data distribution,hinder federated learning's progress.Several approaches have recently emerged for federated learning involving heterogeneous clients with varying computational capabilities(namely,heterogeneous federated learning).However,heterogeneous federated learning faces two key challenges:optimising model size and determining client selection ratios.Moreover,efficiently aggregating local models from clients with diverse capabilities is crucial for addressing system heterogeneity and communication efficiency.This paper proposes an evolutionary multiobjective optimisation framework for heterogeneous federated learning(MOHFL)to address these issues.Our approach elegantly formulates and solves a biobjective optimisation problem that minimises communication cost and model error rate.The decision variables in this framework comprise model sizes and client selection ratios for each Q client cluster,yielding a total of 2×Q optimisation parameters to be tuned.We develop a partition-based strategy for MOHFL that segregates clients into clusters based on their communication and computation capabilities.Additionally,we implement an adaptive model sizing mechanism that dynamically assigns appropriate subnetwork architectures to clients based on their computational constraints.We also propose a unified aggregation framework to combine models of varying sizes from heterogeneous clients effectively.Extensive experiments on multiple datasets demonstrate the effectiveness and superiority of our proposed method compared to existing approaches.展开更多
Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrain...Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrained onboard devices to nearby Roadside Unit(RSU),thereby achieving lower delay and energy consumption.However,due to the limited storage capacity and energy budget of RSUs,it is challenging to meet the demands of the highly dynamic Internet of Vehicles(IoV)environment.Therefore,determining reasonable service caching and computation offloading strategies is crucial.To address this,this paper proposes a joint service caching scheme for cloud-edge collaborative IoV computation offloading.By modeling the dynamic optimization problem using Markov Decision Processes(MDP),the scheme jointly optimizes task delay,energy consumption,load balancing,and privacy entropy to achieve better quality of service.Additionally,a dynamic adaptive multi-objective deep reinforcement learning algorithm is proposed.Each Double Deep Q-Network(DDQN)agent obtains rewards for different objectives based on distinct reward functions and dynamically updates the objective weights by learning the value changes between objectives using Radial Basis Function Networks(RBFN),thereby efficiently approximating the Pareto-optimal decisions for multiple objectives.Extensive experiments demonstrate that the proposed algorithm can better coordinate the three-tier computing resources of cloud,edge,and vehicles.Compared to existing algorithms,the proposed method reduces task delay and energy consumption by 10.64%and 5.1%,respectively.展开更多
Autonomous connected vehicles(ACV)involve advanced control strategies to effectively balance safety,efficiency,energy consumption,and passenger comfort.This research introduces a deep reinforcement learning(DRL)-based...Autonomous connected vehicles(ACV)involve advanced control strategies to effectively balance safety,efficiency,energy consumption,and passenger comfort.This research introduces a deep reinforcement learning(DRL)-based car-following(CF)framework employing the Deep Deterministic Policy Gradient(DDPG)algorithm,which integrates a multi-objective reward function that balances the four goals while maintaining safe policy learning.Utilizing real-world driving data from the highD dataset,the proposed model learns adaptive speed control policies suitable for dynamic traffic scenarios.The performance of the DRL-based model is evaluated against a traditional model predictive control-adaptive cruise control(MPC-ACC)controller.Results show that theDRLmodel significantly enhances safety,achieving zero collisions and a higher average time-to-collision(TTC)of 8.45 s,compared to 5.67 s for MPC and 6.12 s for human drivers.For efficiency,the model demonstrates 89.2% headway compliance and maintains speed tracking errors below 1.2 m/s in 90% of cases.In terms of energy optimization,the proposed approach reduces fuel consumption by 5.4% relative to MPC.Additionally,it enhances passenger comfort by lowering jerk values by 65%,achieving 0.12 m/s3 vs.0.34 m/s3 for human drivers.A multi-objective reward function is integrated to ensure stable policy convergence while simultaneously balancing the four key performance metrics.Moreover,the findings underscore the potential of DRL in advancing autonomous vehicle control,offering a robust and sustainable solution for safer,more efficient,and more comfortable transportation systems.展开更多
The exploration of high-performance materials presents a fundamental challenge in materials science,particularly in predicting properties for materials beyond the known range of target property values(extrapolation).T...The exploration of high-performance materials presents a fundamental challenge in materials science,particularly in predicting properties for materials beyond the known range of target property values(extrapolation).This study formally investigated the interpolation-extrapolation trade-off phenomenon in the prediction capabilities of machine learning(ML)models.A new ML scheme was proposed,featuring a newly developed ML model and forward cross-validation-based hyperparameter optimization,which demonstrated superior extrapolation prediction across multiple materials datasets.Based on this ML scheme,multi-objective optimization was performed to systematically identify lightweight Mg-Zn-Al alloys with both high bulk modulus and high Debye temperature.Subsequently,the designed alloys were validated through density functional theory calculations.Furthermore,a three-category classification strategy was summarized through the dual-driven approach combining domain knowledge and data,emphasizing their synergistic potential for materials discovery.The practical framework developed in this study provides a novel research perspective for exploring high-performance materials.展开更多
In the parallel steering coordination control strategy for path tracking,it is difficult to match the current driver steering model using the fixed parameters with the actual driver,and the designed steering coordinat...In the parallel steering coordination control strategy for path tracking,it is difficult to match the current driver steering model using the fixed parameters with the actual driver,and the designed steering coordination control strategy under a single objective and simple conditions is difficult to adapt to the multi-dimensional state variables’input.In this paper,we propose a deep reinforcement learning algorithm-based multi-objective parallel human-machine steering coordination strategy for path tracking considering driver misoperation and external disturbance.Firstly,the driver steering mathematical model is constructed based on the driver preview characteristics and steering delay response,and the driver characteristic parameters are fitted after collecting the actual driver driving data.Secondly,considering that the vehicle is susceptible to the influence of external disturbances during the driving process,the Tube MPC(Tube Model Predictive Control)based path tracking steering controller is designed based on the vehicle system dynamics error model.After verifying that the driver steering model meets the driver steering operation characteristics,DQN(Deep Q-network),DDPG(Deep Deterministic Policy Gradient)and TD3(Twin Delayed Deep Deterministic Policy Gradient)deep reinforcement learning algorithms are utilized to design a multi-objective parallel steering coordination strategy which satisfies the multi-dimensional state variables’input of the vehicle.Finally,the tracking accuracy,lateral safety,human-machine conflict and driver steering load evaluation index are designed in different driver operation states and different road environments,and the performance of the parallel steering coordination control strategies with different deep reinforcement learning algorithms and fuzzy algorithms are compared by simulations and hardware in the loop experiments.The results show that the parallel steering collaborative strategy based on a deep reinforcement learning algorithm can more effectively assist the driver in tracking the target path under lateral wind interference and driver misoperation,and the TD3-based coordination control strategy has better overall performance.展开更多
Designing compositions and processing of biodegradable magnesium(Mg)alloys to synergistically en-hance mechanical properties and corrosion resistance using conventional trial-and-error method is a challenging task.Thi...Designing compositions and processing of biodegradable magnesium(Mg)alloys to synergistically en-hance mechanical properties and corrosion resistance using conventional trial-and-error method is a challenging task.This study presents a Bayesian optimization(BO)-based multi-objective framework inte-grated with explainable machine learning(ML)to efficiently explore and optimize the high-dimensional design space of biodegradable Mg alloys.Using ultimate tensile strength(UTS),elongation(EL)and cor-rosion potential(E_(corr))as objective properties,the framework balances these conflicting objectives and identifies optimal solutions.A novel biodegradable Mg alloy(Mg-4.6Zn-0.3Y-0.2Mn-0.1Nd-0.1Gd,wt.%)was successfully designed,demonstrating a UTS of 320 MPa,EL of 22%and E_(corr) of−1.60 V(tested in 37℃ simulated body fluid).Compared to JDBM,the UTS has increased by 13 MPa,the EL has improved by 6.1%,and the E_(corr) has risen by 0.02 V.The experimental results presented close agreement with predicted values,validating the proposed framework.The Shapley Additive Explanation method was em-ployed to interpret the ML models,revealing extrusion temperature and Zn content as key parameters driving the optimization design.The strategy provided in this study is universal and offers a potential approach for addressing high-dimensional multi-objective optimization challenges in material develop-ment.展开更多
Achieving the simultaneous enhancement of strength and ductility in laser powder bed fused (LPBF-ed) titanium (Ti) is challenging due to the complex, high-dimensional parameter space and interactions between parameter...Achieving the simultaneous enhancement of strength and ductility in laser powder bed fused (LPBF-ed) titanium (Ti) is challenging due to the complex, high-dimensional parameter space and interactions between parameters and powders. Herein, a hybrid intelligent framework for process parameter optimization of LPBF-ed Ti with improved ultimate tensile strength (UTS) and elongation (EL) was proposed. It combines the data augmentation method (AVG ± EC × SD), the multi-model fusion stacking ensemble learning model (GBDT-BPNN-XGBoost), the interpretable machine learning method and the non-dominated ranking genetic algorithm (NSGA-Ⅱ). The GBDT-BPNN-XGBoost outperforms single models in predicting UTS and EL across the accuracy, generalization ability and stability. The SHAP analysis reveals that laser power (P) is the most important feature affecting both UTS and EL, and it has a positive impact on them when P < 220 W. The UTS and EL of samples fabricated by the optimal process parameters were 718 ± 5 MPa and 27.9 % ± 0.1 %, respectively. The outstanding strength-ductility balance is attributable to the forward stresses in hard α'-martensite and back stresses in soft αm'-martensite induced by the strain gradients of hetero-microstructure. The back stresses strengthen the soft αm'-martensite, improving the overall UTS. The forward stresses stimulate the activation of dislocations in hard α'-martensite and the generation of 〈c + a〉 dislocations, allowing the plastic strain to occur in hard regions and enhancing the overall ductility. This work provides a feasible strategy for multi-objective optimization and valuable insights into tailoring the microstructure for improving mechanical properties.展开更多
Accurate determination of rock mass parameters is essential for ensuring the accuracy of numericalsimulations. Displacement back-analysis is the most widely used method;however, the reliability of thecurrent approache...Accurate determination of rock mass parameters is essential for ensuring the accuracy of numericalsimulations. Displacement back-analysis is the most widely used method;however, the reliability of thecurrent approaches remains unsatisfactory. Therefore, in this paper, a multistage rock mass parameterback-analysis method, that considers the construction process and displacement losses is proposed andimplemented through the coupling of numerical simulation, auto-machine learning (AutoML), andmulti-objective optimization algorithms (MOOAs). First, a parametric modeling platform for mechanizedtwin tunnels is developed, generating a dataset through extensive numerical simulations. Next, theAutoML method is utilized to establish a surrogate model linking rock parameters and displacements.The tunnel construction process is divided into multiple stages, transforming the rock mass parameterback-analysis into a multi-objective optimization problem, for which multi-objective optimization algorithmsare introduced to obtain the rock mass parameters. The newly proposed rock mass parameterback-analysis method is validated in a mechanized twin tunnel project, and its accuracy and effectivenessare demonstrated. Compared with traditional single-stage back-analysis methods, the proposedmodel decreases the average absolute percentage error from 12.73% to 4.34%, significantly improving theaccuracy of the back-analysis. Moreover, although the accuracy of back analysis significantly increaseswith the number of construction stages considered, the back analysis time is acceptable. This studyprovides a new method for displacement back analysis that is efficient and accurate, thereby paving theway for precise parameter determination in numerical simulations.展开更多
This paper investigates a distributed heterogeneous hybrid blocking flow-shop scheduling problem(DHHBFSP)designed to minimize the total tardiness and total energy consumption simultaneously,and proposes an improved pr...This paper investigates a distributed heterogeneous hybrid blocking flow-shop scheduling problem(DHHBFSP)designed to minimize the total tardiness and total energy consumption simultaneously,and proposes an improved proximal policy optimization(IPPO)method to make real-time decisions for the DHHBFSP.A multi-objective Markov decision process is modeled for the DHHBFSP,where the reward function is represented by a vector with dynamic weights instead of the common objectiverelated scalar value.A factory agent(FA)is formulated for each factory to select unscheduled jobs and is trained by the proposed IPPO to improve the decision quality.Multiple FAs work asynchronously to allocate jobs that arrive randomly at the shop.A two-stage training strategy is introduced in the IPPO,which learns from both single-and dual-policy data for better data utilization.The proposed IPPO is tested on randomly generated instances and compared with variants of the basic proximal policy optimization(PPO),dispatch rules,multi-objective metaheuristics,and multi-agent reinforcement learning methods.Extensive experimental results suggest that the proposed strategies offer significant improvements to the basic PPO,and the proposed IPPO outperforms the state-of-the-art scheduling methods in both convergence and solution quality.展开更多
The biomass and coal co-pyrolysis (BCP) technology combines the advantages of both resources, achieving efficient resource complementarity, reducing reliance on coal, and minimizing pollutant emissions. However, this ...The biomass and coal co-pyrolysis (BCP) technology combines the advantages of both resources, achieving efficient resource complementarity, reducing reliance on coal, and minimizing pollutant emissions. However, this process still encounters numerous challenges in attaining optimal economic and environmental performance. Therefore, an ensemble learning (EL) framework is proposed for the BCP process in this study to optimize the synergistic benefits while minimizing negative environmental impacts. Six different ensemble learning models are developed to investigate the impact of input features, such as biomass characteristics, coal characteristics, and pyrolysis conditions on the product profit and CO_(2) emissions of the BCP processes. The Optuna method is further employed to automatically optimize the hyperparameters of BCP process models for enhancing their predictive accuracy and robustness. The results indicate that the categorical boosting (CAB) model of the BCP process has demonstrated exceptional performance in accurately predicting its product profit and CO_(2) emission (R2>0.92) after undergoing five-fold cross-validation. To enhance the interpretability of this preferred model, the Shapley additive explanations and partial dependence plot analyses are conducted to evaluate the impact and importance of biomass characteristics, coal characteristics, and pyrolysis conditions on the product profitability and CO_(2) emissions of the BCP processes. Finally, the preferred model coupled with a reference vector guided evolutionary algorithm is carried to identify the optimal conditions for maximizing the product profit of BCP process products while minimizing CO_(2) emissions. It indicates the optimal BCP process can achieve high product profits (5290.85 CNY·t−1) and low CO_(2) emissions (7.45 kg·t^(−1)).展开更多
The uplift resistance of the soil overlying shield tunnels significantly impacts their anti-floating stability.However,research on uplift resistance concerning special-shaped shield tunnels is limited.This study combi...The uplift resistance of the soil overlying shield tunnels significantly impacts their anti-floating stability.However,research on uplift resistance concerning special-shaped shield tunnels is limited.This study combines numerical simulation with machine learning techniques to explore this issue.It presents a summary of special-shaped tunnel geometries and introduces a shape coefficient.Through the finite element software,Plaxis3D,the study simulates six key parameters—shape coefficient,burial depth ratio,tunnel’s longest horizontal length,internal friction angle,cohesion,and soil submerged bulk density—that impact uplift resistance across different conditions.Employing XGBoost and ANN methods,the feature importance of each parameter was analyzed based on the numerical simulation results.The findings demonstrate that a tunnel shape more closely resembling a circle leads to reduced uplift resistance in the overlying soil,whereas other parameters exhibit the contrary effects.Furthermore,the study reveals a diminishing trend in the feature importance of buried depth ratio,internal friction angle,tunnel longest horizontal length,cohesion,soil submerged bulk density,and shape coefficient in influencing uplift resistance.展开更多
Countries around the world have been making efforts to reduce pollutant emissions. However, the response of global black carbon(BC) aging to emission changes remains unclear. Using the Community Atmosphere Model versi...Countries around the world have been making efforts to reduce pollutant emissions. However, the response of global black carbon(BC) aging to emission changes remains unclear. Using the Community Atmosphere Model version 6 with a machine-learning-integrated four-mode version of the Modal Aerosol Module, we quantify global BC aging responses to emission reductions for 2011–2018 and for 2050 and 2100 under carbon neutrality. During 2011–18, global trends in BC aging degree(mass ratio of coatings to BC, R_(BC)) exhibited marked regional disparities, with a significant increase in China(5.4% yr^(-1)), which contrasts with minimal changes in the USA, Europe, and India. The divergence is attributed to opposing trends in secondary organic aerosol(SOA) and sulfate coatings, driven by regional changes in the emission ratios of corresponding coating precursors to BC(volatile organic compounds-VOCs/BC and SO_(2)/BC). Projections under carbon neutrality reveal that R_(BC) will increase globally by 47%(118%) in 2050(2100), with strong convergent increases expected across major source regions. The R_(BC) increase, primarily driven by enhanced SOA coatings due to sharper BC reductions relative to VOCs, will enhance the global BC mass absorption cross-section(MAC) by 11%(17%) in 2050(2100).Consequently, although the global BC burden will decline sharply by 60%(76%), the enhanced MAC partially offsets the magnitude of the decline in the BC direct radiative effect, resulting in the moderation of global BC DRE decreases to 88%(92%) of the BC burden reductions in 2050(2100). This study highlights the globally enhanced BC aging and light absorption capacity under carbon neutrality, thereby partly offsetting the impact of BC direct emission reductions on future changes in BC radiative effects globally.展开更多
The demand for extended electric vehicle(EV)range necessitates advanced lightweighting strategies.This study introduces a materials genome approach,augmented by machine learning(ML),for optimizing lightweight composit...The demand for extended electric vehicle(EV)range necessitates advanced lightweighting strategies.This study introduces a materials genome approach,augmented by machine learning(ML),for optimizing lightweight composite designs for EVs.A comprehensive materials genome database was developed,encompassing composites based on carbon,glass,and natural fibers.This database systematically records critical parameters such as mechanical properties,density,cost,and environmental impact.Machine learning models,including Random Forest,Support Vector Machines,and Artificial Neural Networks,were employed to construct a predictive system for material performance.Subsequent material composition optimization was performed using amulti-objective genetic algorithm.Experimental validation demonstrated that an optimized carbon fiber/bio-based resin composite achieved a 45%weight reduction compared to conventional steel,while maintaining equivalent structural strength.The predictive accuracy of the models reached 94.2%.A cost-benefit analysis indicated that despite a 15%increase in material cost,the overall vehicle energy consumption decreased by 12%,leading to an 18%total cost saving over a five-year operational lifecycle,under a representative mid-size battery electric vehicle(BEV)operational scenario.展开更多
Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challeng...Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challenge due to factors such as light scattering,absorption,restricted visibility,and ambient noise.The advancement of deep learning has introduced powerful techniques for processing large amounts of unstructured and imperfect data collected from underwater environments.This study evaluated the efficacy of the You Only Look Once(YOLO)algorithm,a real-time object detection and localization model based on convolutional neural networks,in identifying and classifying various types of pipeline defects in underwater settings.YOLOv8,the latest evolution in the YOLO family,integrates advanced capabilities,such as anchor-free detection,a cross-stage partial network backbone for efficient feature extraction,and a feature pyramid network+path aggregation network neck for robust multi-scale object detection,which make it particularly well-suited for complex underwater environments.Due to the lack of suitable open-access datasets for underwater pipeline defects,a custom dataset was captured using a remotely operated vehicle in a controlled environment.This application has the following assets available for use.Extensive experimentation demonstrated that YOLOv8 X-Large consistently outperformed other models in terms of pipe defect detection and classification and achieved a strong balance between precision and recall in identifying pipeline cracks,rust,corners,defective welds,flanges,tapes,and holes.This research establishes the baseline performance of YOLOv8 for underwater defect detection and showcases its potential to enhance the reliability and efficiency of pipeline inspection tasks in challenging underwater environments.展开更多
Cyclohexene is an important raw material in the production of nylon.Selective hydrogenation of benzene is a key method for preparing cyclohexene.However,the Ru catalysts used in current industrial processes still face...Cyclohexene is an important raw material in the production of nylon.Selective hydrogenation of benzene is a key method for preparing cyclohexene.However,the Ru catalysts used in current industrial processes still face challenges,including high metal usage,high process costs,and low cyclohexene yield.This study utilizes existing literature data combined with machine learning methods to analyze the factors influencing benzene conversion,cyclohexene selectivity,and yield in the benzene hydrogenation to cyclohexene reaction.It constructs predictive models based on XGBoost and Random Forest algorithms.After analysis,it was found that reaction time,Ru content,and space velocity are key factors influencing cyclohexene yield,selectivity,and benzene conversion.Shapley Additive Explanations(SHAP)analysis and feature importance analysis further revealed the contribution of each variable to the reaction outcomes.Additionally,we randomly generated one million variable combinations using the Dirichlet distribution to attempt to predict high-yield catalyst formulations.This paper provides new insights into the application of machine learning in heterogeneous catalysis and offers some reference for further research.展开更多
The electrocatalytic reduction of nitric oxide for ammonia synthesis(NORR)is a key green energy conversion technology.Its efficiency relies on high-performance electrocatalysts to enhance both ammonia yield(Y_(NH3))an...The electrocatalytic reduction of nitric oxide for ammonia synthesis(NORR)is a key green energy conversion technology.Its efficiency relies on high-performance electrocatalysts to enhance both ammonia yield(Y_(NH3))and Faradaic efficiency(F_(NH3)).However,conventional experimental methods for screening high-activity NORR catalysts often entail high resource consumption and time costs.Machine learning combined with SHAP feature analysis was employed to establish a stacked ensemble model that integrates multiple algorithms,to allow for a systematic investigation of the key descriptors governing NORR performance based on an experimental dataset.Evaluation of eight model algorithms revealed that the Stacked-SVR model achieved an R^(2)of 0.9223 and an RMSE of 0.0608 for predicting on the test set,whereas the Stacked-RF model achieved an R^(2)of 0.9042 and an RMSE of 0.0900 for predicting.The stacked ensemble model integrates the strengths of individual algorithms and demonstrates strong NORR prediction performance while avoiding overfitting.SHAP feature analysis results revealed that the Cu content in the catalyst composition has the most significant impact on catalytic performance.Moreover,the combination of the wet chemical reduction synthesis,a carbon fiber(CF)conductive substrate,and HCl electrolyte is more favorable for enhancing catalytic activity.Additionally,moderately lowering the working potential,controlling the electrolyte volume at low to medium levels,reducing catalyst loading,and increasing electrolyte concentration were found to synergistically enhance both and.展开更多
Federated Learning(FL)has become a leading decentralized solution that enables multiple clients to train a model in a collaborative environment without directly sharing raw data,making it suitable for privacy-sensitiv...Federated Learning(FL)has become a leading decentralized solution that enables multiple clients to train a model in a collaborative environment without directly sharing raw data,making it suitable for privacy-sensitive applications such as healthcare,finance,and smart systems.As the field continues to evolve,the research field has become more complex and scattered,covering different system designs,training methods,and privacy techniques.This survey is organized around the three core challenges:how the data is distributed,how models are synchronized,and how to defend against attacks.It provides a structured and up-to-date review of FL research from 2023 to 2025,offering a unified taxonomy that categorizes works by data distribution(Horizontal FL,Vertical FL,Federated Transfer Learning,and Personalized FL),training synchronization(synchronous and asynchronous FL),optimization strategies,and threat models(data leakage and poisoning attacks).In particular,we summarize the latest contributions in Vertical FL frameworks for secure multi-party learning,communication-efficient Horizontal FL,and domain-adaptive Federated Transfer Learning.Furthermore,we examine synchronization techniques addressing system heterogeneity,including straggler mitigation in synchronous FL and staleness management in asynchronous FL.The survey covers security threats in FL,such as gradient inversion,membership inference,and poisoning attacks,as well as their defense strategies that include privacy-preserving aggregation and anomaly detection.The paper concludes by outlining unresolved issues and highlighting challenges in handling personalized models,scalability,and real-world adoption.展开更多
Split Learning(SL)has been promoted as a promising collaborative machine learning technique designed to address data privacy and resource efficiency.Specifically,neural networks are divided into client and server subn...Split Learning(SL)has been promoted as a promising collaborative machine learning technique designed to address data privacy and resource efficiency.Specifically,neural networks are divided into client and server subnetworks in order to mitigate the exposure of sensitive data and reduce the overhead on client devices,thereby making SL particularly suitable for resource-constrained devices.Although SL prevents the direct transmission of raw data,it does not alleviate entirely the risk of privacy breaches.In fact,the data intermediately transmitted to the server sub-model may include patterns or information that could reveal sensitive data.Moreover,achieving a balance between model utility and data privacy has emerged as a challenging problem.In this article,we propose a novel defense approach that combines:(i)Adversarial learning,and(ii)Network channel pruning.In particular,the proposed adversarial learning approach is specifically designed to reduce the risk of private data exposure while maintaining high performance for the utility task.On the other hand,the suggested channel pruning enables the model to adaptively adjust and reactivate pruned channels while conducting adversarial training.The integration of these two techniques reduces the informativeness of the intermediate data transmitted by the client sub-model,thereby enhancing its robustness against attribute inference attacks without adding significant computational overhead,making it wellsuited for IoT devices,mobile platforms,and Internet of Vehicles(IoV)scenarios.The proposed defense approach was evaluated using EfficientNet-B0,a widely adopted compact model,along with three benchmark datasets.The obtained results showcased its superior defense capability against attribute inference attacks compared to existing state-of-the-art methods.This research’s findings demonstrated the effectiveness of the proposed channel pruning-based adversarial training approach in achieving the intended compromise between utility and privacy within SL frameworks.In fact,the classification accuracy attained by the attackers witnessed a drastic decrease of 70%.展开更多
基金supported by the National Research Foundation of Korea grant funded by the Korea government(RS-2023-00217116)。
文摘Federated learning is a distributed framework that trains a centralised model using data from multiple clients without transferring that data to a central server.Despite rapid progress,federated learning still faces several unsolved challenges.Specifically,communication costs and system heterogeneity,such as nonidentical data distribution,hinder federated learning's progress.Several approaches have recently emerged for federated learning involving heterogeneous clients with varying computational capabilities(namely,heterogeneous federated learning).However,heterogeneous federated learning faces two key challenges:optimising model size and determining client selection ratios.Moreover,efficiently aggregating local models from clients with diverse capabilities is crucial for addressing system heterogeneity and communication efficiency.This paper proposes an evolutionary multiobjective optimisation framework for heterogeneous federated learning(MOHFL)to address these issues.Our approach elegantly formulates and solves a biobjective optimisation problem that minimises communication cost and model error rate.The decision variables in this framework comprise model sizes and client selection ratios for each Q client cluster,yielding a total of 2×Q optimisation parameters to be tuned.We develop a partition-based strategy for MOHFL that segregates clients into clusters based on their communication and computation capabilities.Additionally,we implement an adaptive model sizing mechanism that dynamically assigns appropriate subnetwork architectures to clients based on their computational constraints.We also propose a unified aggregation framework to combine models of varying sizes from heterogeneous clients effectively.Extensive experiments on multiple datasets demonstrate the effectiveness and superiority of our proposed method compared to existing approaches.
基金supported by Key Science and Technology Program of Henan Province,China(Grant Nos.242102210147,242102210027)Fujian Province Young and Middle aged Teacher Education Research Project(Science and Technology Category)(No.JZ240101)(Corresponding author:Dong Yuan).
文摘Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrained onboard devices to nearby Roadside Unit(RSU),thereby achieving lower delay and energy consumption.However,due to the limited storage capacity and energy budget of RSUs,it is challenging to meet the demands of the highly dynamic Internet of Vehicles(IoV)environment.Therefore,determining reasonable service caching and computation offloading strategies is crucial.To address this,this paper proposes a joint service caching scheme for cloud-edge collaborative IoV computation offloading.By modeling the dynamic optimization problem using Markov Decision Processes(MDP),the scheme jointly optimizes task delay,energy consumption,load balancing,and privacy entropy to achieve better quality of service.Additionally,a dynamic adaptive multi-objective deep reinforcement learning algorithm is proposed.Each Double Deep Q-Network(DDQN)agent obtains rewards for different objectives based on distinct reward functions and dynamically updates the objective weights by learning the value changes between objectives using Radial Basis Function Networks(RBFN),thereby efficiently approximating the Pareto-optimal decisions for multiple objectives.Extensive experiments demonstrate that the proposed algorithm can better coordinate the three-tier computing resources of cloud,edge,and vehicles.Compared to existing algorithms,the proposed method reduces task delay and energy consumption by 10.64%and 5.1%,respectively.
基金the Hebei Province Science and Technology Plan Project(19221909D)rincess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R308),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Autonomous connected vehicles(ACV)involve advanced control strategies to effectively balance safety,efficiency,energy consumption,and passenger comfort.This research introduces a deep reinforcement learning(DRL)-based car-following(CF)framework employing the Deep Deterministic Policy Gradient(DDPG)algorithm,which integrates a multi-objective reward function that balances the four goals while maintaining safe policy learning.Utilizing real-world driving data from the highD dataset,the proposed model learns adaptive speed control policies suitable for dynamic traffic scenarios.The performance of the DRL-based model is evaluated against a traditional model predictive control-adaptive cruise control(MPC-ACC)controller.Results show that theDRLmodel significantly enhances safety,achieving zero collisions and a higher average time-to-collision(TTC)of 8.45 s,compared to 5.67 s for MPC and 6.12 s for human drivers.For efficiency,the model demonstrates 89.2% headway compliance and maintains speed tracking errors below 1.2 m/s in 90% of cases.In terms of energy optimization,the proposed approach reduces fuel consumption by 5.4% relative to MPC.Additionally,it enhances passenger comfort by lowering jerk values by 65%,achieving 0.12 m/s3 vs.0.34 m/s3 for human drivers.A multi-objective reward function is integrated to ensure stable policy convergence while simultaneously balancing the four key performance metrics.Moreover,the findings underscore the potential of DRL in advancing autonomous vehicle control,offering a robust and sustainable solution for safer,more efficient,and more comfortable transportation systems.
基金supported by National Natural Science Foundation of China(No.51671075 and 51971086)Natural Science Foundation of Heilongjiang Province of China(No.LH2022E081).
文摘The exploration of high-performance materials presents a fundamental challenge in materials science,particularly in predicting properties for materials beyond the known range of target property values(extrapolation).This study formally investigated the interpolation-extrapolation trade-off phenomenon in the prediction capabilities of machine learning(ML)models.A new ML scheme was proposed,featuring a newly developed ML model and forward cross-validation-based hyperparameter optimization,which demonstrated superior extrapolation prediction across multiple materials datasets.Based on this ML scheme,multi-objective optimization was performed to systematically identify lightweight Mg-Zn-Al alloys with both high bulk modulus and high Debye temperature.Subsequently,the designed alloys were validated through density functional theory calculations.Furthermore,a three-category classification strategy was summarized through the dual-driven approach combining domain knowledge and data,emphasizing their synergistic potential for materials discovery.The practical framework developed in this study provides a novel research perspective for exploring high-performance materials.
基金Supported by National Natural Science Foundation of China(Grant Nos.U22A20246,52372382)Hefei Municipal Natural Science Foundation(Grant No.2022008)+1 种基金the Open Fund of State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures(Grant No.KF2023-06)S&T Program of Hebei(Grant No.225676162GH).
文摘In the parallel steering coordination control strategy for path tracking,it is difficult to match the current driver steering model using the fixed parameters with the actual driver,and the designed steering coordination control strategy under a single objective and simple conditions is difficult to adapt to the multi-dimensional state variables’input.In this paper,we propose a deep reinforcement learning algorithm-based multi-objective parallel human-machine steering coordination strategy for path tracking considering driver misoperation and external disturbance.Firstly,the driver steering mathematical model is constructed based on the driver preview characteristics and steering delay response,and the driver characteristic parameters are fitted after collecting the actual driver driving data.Secondly,considering that the vehicle is susceptible to the influence of external disturbances during the driving process,the Tube MPC(Tube Model Predictive Control)based path tracking steering controller is designed based on the vehicle system dynamics error model.After verifying that the driver steering model meets the driver steering operation characteristics,DQN(Deep Q-network),DDPG(Deep Deterministic Policy Gradient)and TD3(Twin Delayed Deep Deterministic Policy Gradient)deep reinforcement learning algorithms are utilized to design a multi-objective parallel steering coordination strategy which satisfies the multi-dimensional state variables’input of the vehicle.Finally,the tracking accuracy,lateral safety,human-machine conflict and driver steering load evaluation index are designed in different driver operation states and different road environments,and the performance of the parallel steering coordination control strategies with different deep reinforcement learning algorithms and fuzzy algorithms are compared by simulations and hardware in the loop experiments.The results show that the parallel steering collaborative strategy based on a deep reinforcement learning algorithm can more effectively assist the driver in tracking the target path under lateral wind interference and driver misoperation,and the TD3-based coordination control strategy has better overall performance.
基金financially supported by the National Natu-ral Science Foundation of China(No.52301133)the China Post-doctoral Science Foundation(No.2023M730276)+1 种基金the Young Elite Scientists Sponsorship Program by China Association for Science and Technology(No.YESS20210415)the Graduate Innovation Pro-gram of Chongqing University of Science and Technology(No.YKJCX2320218).
文摘Designing compositions and processing of biodegradable magnesium(Mg)alloys to synergistically en-hance mechanical properties and corrosion resistance using conventional trial-and-error method is a challenging task.This study presents a Bayesian optimization(BO)-based multi-objective framework inte-grated with explainable machine learning(ML)to efficiently explore and optimize the high-dimensional design space of biodegradable Mg alloys.Using ultimate tensile strength(UTS),elongation(EL)and cor-rosion potential(E_(corr))as objective properties,the framework balances these conflicting objectives and identifies optimal solutions.A novel biodegradable Mg alloy(Mg-4.6Zn-0.3Y-0.2Mn-0.1Nd-0.1Gd,wt.%)was successfully designed,demonstrating a UTS of 320 MPa,EL of 22%and E_(corr) of−1.60 V(tested in 37℃ simulated body fluid).Compared to JDBM,the UTS has increased by 13 MPa,the EL has improved by 6.1%,and the E_(corr) has risen by 0.02 V.The experimental results presented close agreement with predicted values,validating the proposed framework.The Shapley Additive Explanation method was em-ployed to interpret the ML models,revealing extrusion temperature and Zn content as key parameters driving the optimization design.The strategy provided in this study is universal and offers a potential approach for addressing high-dimensional multi-objective optimization challenges in material develop-ment.
基金supported by the National Natural Sci-ence Foundation of China(Nos.52274359 and 52304379)the China National Postdoctoral Program for Innovative Talents(No.BX20220034)+2 种基金the China Postdoctoral Science Foundation(No.2022M720403)the AECC University Research Cooperation Project(No.HFZL2021CXY021)the Interdisciplinary Research Project for Young Teachers of USTB(Fundamental Research Funds for the Central Universities)(No.FRF-IDRY-23-025).
文摘Achieving the simultaneous enhancement of strength and ductility in laser powder bed fused (LPBF-ed) titanium (Ti) is challenging due to the complex, high-dimensional parameter space and interactions between parameters and powders. Herein, a hybrid intelligent framework for process parameter optimization of LPBF-ed Ti with improved ultimate tensile strength (UTS) and elongation (EL) was proposed. It combines the data augmentation method (AVG ± EC × SD), the multi-model fusion stacking ensemble learning model (GBDT-BPNN-XGBoost), the interpretable machine learning method and the non-dominated ranking genetic algorithm (NSGA-Ⅱ). The GBDT-BPNN-XGBoost outperforms single models in predicting UTS and EL across the accuracy, generalization ability and stability. The SHAP analysis reveals that laser power (P) is the most important feature affecting both UTS and EL, and it has a positive impact on them when P < 220 W. The UTS and EL of samples fabricated by the optimal process parameters were 718 ± 5 MPa and 27.9 % ± 0.1 %, respectively. The outstanding strength-ductility balance is attributable to the forward stresses in hard α'-martensite and back stresses in soft αm'-martensite induced by the strain gradients of hetero-microstructure. The back stresses strengthen the soft αm'-martensite, improving the overall UTS. The forward stresses stimulate the activation of dislocations in hard α'-martensite and the generation of 〈c + a〉 dislocations, allowing the plastic strain to occur in hard regions and enhancing the overall ductility. This work provides a feasible strategy for multi-objective optimization and valuable insights into tailoring the microstructure for improving mechanical properties.
基金supported by the National Natural Science Foundation of China(Grant Nos.52090081,52079068)the State Key Laboratory of Hydroscience and Hydraulic Engineering(Grant No.2021-KY-04).
文摘Accurate determination of rock mass parameters is essential for ensuring the accuracy of numericalsimulations. Displacement back-analysis is the most widely used method;however, the reliability of thecurrent approaches remains unsatisfactory. Therefore, in this paper, a multistage rock mass parameterback-analysis method, that considers the construction process and displacement losses is proposed andimplemented through the coupling of numerical simulation, auto-machine learning (AutoML), andmulti-objective optimization algorithms (MOOAs). First, a parametric modeling platform for mechanizedtwin tunnels is developed, generating a dataset through extensive numerical simulations. Next, theAutoML method is utilized to establish a surrogate model linking rock parameters and displacements.The tunnel construction process is divided into multiple stages, transforming the rock mass parameterback-analysis into a multi-objective optimization problem, for which multi-objective optimization algorithmsare introduced to obtain the rock mass parameters. The newly proposed rock mass parameterback-analysis method is validated in a mechanized twin tunnel project, and its accuracy and effectivenessare demonstrated. Compared with traditional single-stage back-analysis methods, the proposedmodel decreases the average absolute percentage error from 12.73% to 4.34%, significantly improving theaccuracy of the back-analysis. Moreover, although the accuracy of back analysis significantly increaseswith the number of construction stages considered, the back analysis time is acceptable. This studyprovides a new method for displacement back analysis that is efficient and accurate, thereby paving theway for precise parameter determination in numerical simulations.
基金partially supported by the National Key Research and Development Program of the Ministry of Science and Technology of China(2022YFE0114200)the National Natural Science Foundation of China(U20A6004).
文摘This paper investigates a distributed heterogeneous hybrid blocking flow-shop scheduling problem(DHHBFSP)designed to minimize the total tardiness and total energy consumption simultaneously,and proposes an improved proximal policy optimization(IPPO)method to make real-time decisions for the DHHBFSP.A multi-objective Markov decision process is modeled for the DHHBFSP,where the reward function is represented by a vector with dynamic weights instead of the common objectiverelated scalar value.A factory agent(FA)is formulated for each factory to select unscheduled jobs and is trained by the proposed IPPO to improve the decision quality.Multiple FAs work asynchronously to allocate jobs that arrive randomly at the shop.A two-stage training strategy is introduced in the IPPO,which learns from both single-and dual-policy data for better data utilization.The proposed IPPO is tested on randomly generated instances and compared with variants of the basic proximal policy optimization(PPO),dispatch rules,multi-objective metaheuristics,and multi-agent reinforcement learning methods.Extensive experimental results suggest that the proposed strategies offer significant improvements to the basic PPO,and the proposed IPPO outperforms the state-of-the-art scheduling methods in both convergence and solution quality.
基金support from the National Natural Science Foundation of China(22108052).
文摘The biomass and coal co-pyrolysis (BCP) technology combines the advantages of both resources, achieving efficient resource complementarity, reducing reliance on coal, and minimizing pollutant emissions. However, this process still encounters numerous challenges in attaining optimal economic and environmental performance. Therefore, an ensemble learning (EL) framework is proposed for the BCP process in this study to optimize the synergistic benefits while minimizing negative environmental impacts. Six different ensemble learning models are developed to investigate the impact of input features, such as biomass characteristics, coal characteristics, and pyrolysis conditions on the product profit and CO_(2) emissions of the BCP processes. The Optuna method is further employed to automatically optimize the hyperparameters of BCP process models for enhancing their predictive accuracy and robustness. The results indicate that the categorical boosting (CAB) model of the BCP process has demonstrated exceptional performance in accurately predicting its product profit and CO_(2) emission (R2>0.92) after undergoing five-fold cross-validation. To enhance the interpretability of this preferred model, the Shapley additive explanations and partial dependence plot analyses are conducted to evaluate the impact and importance of biomass characteristics, coal characteristics, and pyrolysis conditions on the product profitability and CO_(2) emissions of the BCP processes. Finally, the preferred model coupled with a reference vector guided evolutionary algorithm is carried to identify the optimal conditions for maximizing the product profit of BCP process products while minimizing CO_(2) emissions. It indicates the optimal BCP process can achieve high product profits (5290.85 CNY·t−1) and low CO_(2) emissions (7.45 kg·t^(−1)).
基金Guangzhou Metro Scientific Research Project(No.JT204-100111-23001)Chongqing Municipal Special Project for Technological Innovation and Application Development(No.CSTB2022TIAD-KPX0101)Science and Technology Research and Development Program of China State Railway Group Co.,Ltd.(No.N2023G045)。
文摘The uplift resistance of the soil overlying shield tunnels significantly impacts their anti-floating stability.However,research on uplift resistance concerning special-shaped shield tunnels is limited.This study combines numerical simulation with machine learning techniques to explore this issue.It presents a summary of special-shaped tunnel geometries and introduces a shape coefficient.Through the finite element software,Plaxis3D,the study simulates six key parameters—shape coefficient,burial depth ratio,tunnel’s longest horizontal length,internal friction angle,cohesion,and soil submerged bulk density—that impact uplift resistance across different conditions.Employing XGBoost and ANN methods,the feature importance of each parameter was analyzed based on the numerical simulation results.The findings demonstrate that a tunnel shape more closely resembling a circle leads to reduced uplift resistance in the overlying soil,whereas other parameters exhibit the contrary effects.Furthermore,the study reveals a diminishing trend in the feature importance of buried depth ratio,internal friction angle,tunnel longest horizontal length,cohesion,soil submerged bulk density,and shape coefficient in influencing uplift resistance.
基金supported by the National Natural Science Foundation of China (42505149,41925023,U2342223,42105069,and 91744208)the China Postdoctoral Science Foundation (2025M770303)+1 种基金the Fundamental Research Funds for the Central Universities (14380230)the Jiangsu Funding Program for Excellent Postdoctoral Talent,and Jiangsu Collaborative Innovation Center of Climate Change。
文摘Countries around the world have been making efforts to reduce pollutant emissions. However, the response of global black carbon(BC) aging to emission changes remains unclear. Using the Community Atmosphere Model version 6 with a machine-learning-integrated four-mode version of the Modal Aerosol Module, we quantify global BC aging responses to emission reductions for 2011–2018 and for 2050 and 2100 under carbon neutrality. During 2011–18, global trends in BC aging degree(mass ratio of coatings to BC, R_(BC)) exhibited marked regional disparities, with a significant increase in China(5.4% yr^(-1)), which contrasts with minimal changes in the USA, Europe, and India. The divergence is attributed to opposing trends in secondary organic aerosol(SOA) and sulfate coatings, driven by regional changes in the emission ratios of corresponding coating precursors to BC(volatile organic compounds-VOCs/BC and SO_(2)/BC). Projections under carbon neutrality reveal that R_(BC) will increase globally by 47%(118%) in 2050(2100), with strong convergent increases expected across major source regions. The R_(BC) increase, primarily driven by enhanced SOA coatings due to sharper BC reductions relative to VOCs, will enhance the global BC mass absorption cross-section(MAC) by 11%(17%) in 2050(2100).Consequently, although the global BC burden will decline sharply by 60%(76%), the enhanced MAC partially offsets the magnitude of the decline in the BC direct radiative effect, resulting in the moderation of global BC DRE decreases to 88%(92%) of the BC burden reductions in 2050(2100). This study highlights the globally enhanced BC aging and light absorption capacity under carbon neutrality, thereby partly offsetting the impact of BC direct emission reductions on future changes in BC radiative effects globally.
文摘The demand for extended electric vehicle(EV)range necessitates advanced lightweighting strategies.This study introduces a materials genome approach,augmented by machine learning(ML),for optimizing lightweight composite designs for EVs.A comprehensive materials genome database was developed,encompassing composites based on carbon,glass,and natural fibers.This database systematically records critical parameters such as mechanical properties,density,cost,and environmental impact.Machine learning models,including Random Forest,Support Vector Machines,and Artificial Neural Networks,were employed to construct a predictive system for material performance.Subsequent material composition optimization was performed using amulti-objective genetic algorithm.Experimental validation demonstrated that an optimized carbon fiber/bio-based resin composite achieved a 45%weight reduction compared to conventional steel,while maintaining equivalent structural strength.The predictive accuracy of the models reached 94.2%.A cost-benefit analysis indicated that despite a 15%increase in material cost,the overall vehicle energy consumption decreased by 12%,leading to an 18%total cost saving over a five-year operational lifecycle,under a representative mid-size battery electric vehicle(BEV)operational scenario.
文摘Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challenge due to factors such as light scattering,absorption,restricted visibility,and ambient noise.The advancement of deep learning has introduced powerful techniques for processing large amounts of unstructured and imperfect data collected from underwater environments.This study evaluated the efficacy of the You Only Look Once(YOLO)algorithm,a real-time object detection and localization model based on convolutional neural networks,in identifying and classifying various types of pipeline defects in underwater settings.YOLOv8,the latest evolution in the YOLO family,integrates advanced capabilities,such as anchor-free detection,a cross-stage partial network backbone for efficient feature extraction,and a feature pyramid network+path aggregation network neck for robust multi-scale object detection,which make it particularly well-suited for complex underwater environments.Due to the lack of suitable open-access datasets for underwater pipeline defects,a custom dataset was captured using a remotely operated vehicle in a controlled environment.This application has the following assets available for use.Extensive experimentation demonstrated that YOLOv8 X-Large consistently outperformed other models in terms of pipe defect detection and classification and achieved a strong balance between precision and recall in identifying pipeline cracks,rust,corners,defective welds,flanges,tapes,and holes.This research establishes the baseline performance of YOLOv8 for underwater defect detection and showcases its potential to enhance the reliability and efficiency of pipeline inspection tasks in challenging underwater environments.
基金Supported by CAS Basic and Interdisciplinary Frontier Scientific Research Pilot Project(XDB1190300,XDB1190302)Youth Innovation Promotion Association CAS(Y2021056)+1 种基金Joint Fund of the Yulin University and the Dalian National Laboratory for Clean Energy(YLU-DNL Fund 2022007)The special fund for Science and Technology Innovation Teams of Shanxi Province(202304051001007)。
文摘Cyclohexene is an important raw material in the production of nylon.Selective hydrogenation of benzene is a key method for preparing cyclohexene.However,the Ru catalysts used in current industrial processes still face challenges,including high metal usage,high process costs,and low cyclohexene yield.This study utilizes existing literature data combined with machine learning methods to analyze the factors influencing benzene conversion,cyclohexene selectivity,and yield in the benzene hydrogenation to cyclohexene reaction.It constructs predictive models based on XGBoost and Random Forest algorithms.After analysis,it was found that reaction time,Ru content,and space velocity are key factors influencing cyclohexene yield,selectivity,and benzene conversion.Shapley Additive Explanations(SHAP)analysis and feature importance analysis further revealed the contribution of each variable to the reaction outcomes.Additionally,we randomly generated one million variable combinations using the Dirichlet distribution to attempt to predict high-yield catalyst formulations.This paper provides new insights into the application of machine learning in heterogeneous catalysis and offers some reference for further research.
文摘The electrocatalytic reduction of nitric oxide for ammonia synthesis(NORR)is a key green energy conversion technology.Its efficiency relies on high-performance electrocatalysts to enhance both ammonia yield(Y_(NH3))and Faradaic efficiency(F_(NH3)).However,conventional experimental methods for screening high-activity NORR catalysts often entail high resource consumption and time costs.Machine learning combined with SHAP feature analysis was employed to establish a stacked ensemble model that integrates multiple algorithms,to allow for a systematic investigation of the key descriptors governing NORR performance based on an experimental dataset.Evaluation of eight model algorithms revealed that the Stacked-SVR model achieved an R^(2)of 0.9223 and an RMSE of 0.0608 for predicting on the test set,whereas the Stacked-RF model achieved an R^(2)of 0.9042 and an RMSE of 0.0900 for predicting.The stacked ensemble model integrates the strengths of individual algorithms and demonstrates strong NORR prediction performance while avoiding overfitting.SHAP feature analysis results revealed that the Cu content in the catalyst composition has the most significant impact on catalytic performance.Moreover,the combination of the wet chemical reduction synthesis,a carbon fiber(CF)conductive substrate,and HCl electrolyte is more favorable for enhancing catalytic activity.Additionally,moderately lowering the working potential,controlling the electrolyte volume at low to medium levels,reducing catalyst loading,and increasing electrolyte concentration were found to synergistically enhance both and.
文摘Federated Learning(FL)has become a leading decentralized solution that enables multiple clients to train a model in a collaborative environment without directly sharing raw data,making it suitable for privacy-sensitive applications such as healthcare,finance,and smart systems.As the field continues to evolve,the research field has become more complex and scattered,covering different system designs,training methods,and privacy techniques.This survey is organized around the three core challenges:how the data is distributed,how models are synchronized,and how to defend against attacks.It provides a structured and up-to-date review of FL research from 2023 to 2025,offering a unified taxonomy that categorizes works by data distribution(Horizontal FL,Vertical FL,Federated Transfer Learning,and Personalized FL),training synchronization(synchronous and asynchronous FL),optimization strategies,and threat models(data leakage and poisoning attacks).In particular,we summarize the latest contributions in Vertical FL frameworks for secure multi-party learning,communication-efficient Horizontal FL,and domain-adaptive Federated Transfer Learning.Furthermore,we examine synchronization techniques addressing system heterogeneity,including straggler mitigation in synchronous FL and staleness management in asynchronous FL.The survey covers security threats in FL,such as gradient inversion,membership inference,and poisoning attacks,as well as their defense strategies that include privacy-preserving aggregation and anomaly detection.The paper concludes by outlining unresolved issues and highlighting challenges in handling personalized models,scalability,and real-world adoption.
基金supported by a grant(No.CRPG-25-2054)under the Cybersecurity Research and Innovation Pioneers Initiative,provided by the National Cybersecurity Authority(NCA)in the Kingdom of Saudi Arabia.
文摘Split Learning(SL)has been promoted as a promising collaborative machine learning technique designed to address data privacy and resource efficiency.Specifically,neural networks are divided into client and server subnetworks in order to mitigate the exposure of sensitive data and reduce the overhead on client devices,thereby making SL particularly suitable for resource-constrained devices.Although SL prevents the direct transmission of raw data,it does not alleviate entirely the risk of privacy breaches.In fact,the data intermediately transmitted to the server sub-model may include patterns or information that could reveal sensitive data.Moreover,achieving a balance between model utility and data privacy has emerged as a challenging problem.In this article,we propose a novel defense approach that combines:(i)Adversarial learning,and(ii)Network channel pruning.In particular,the proposed adversarial learning approach is specifically designed to reduce the risk of private data exposure while maintaining high performance for the utility task.On the other hand,the suggested channel pruning enables the model to adaptively adjust and reactivate pruned channels while conducting adversarial training.The integration of these two techniques reduces the informativeness of the intermediate data transmitted by the client sub-model,thereby enhancing its robustness against attribute inference attacks without adding significant computational overhead,making it wellsuited for IoT devices,mobile platforms,and Internet of Vehicles(IoV)scenarios.The proposed defense approach was evaluated using EfficientNet-B0,a widely adopted compact model,along with three benchmark datasets.The obtained results showcased its superior defense capability against attribute inference attacks compared to existing state-of-the-art methods.This research’s findings demonstrated the effectiveness of the proposed channel pruning-based adversarial training approach in achieving the intended compromise between utility and privacy within SL frameworks.In fact,the classification accuracy attained by the attackers witnessed a drastic decrease of 70%.
