Early identification and treatment of stroke can greatly improve patient outcomes and quality of life.Although clinical tests such as the Cincinnati Pre-hospital Stroke Scale(CPSS)and the Face Arm Speech Test(FAST)are...Early identification and treatment of stroke can greatly improve patient outcomes and quality of life.Although clinical tests such as the Cincinnati Pre-hospital Stroke Scale(CPSS)and the Face Arm Speech Test(FAST)are commonly used for stroke screening,accurate administration is dependent on specialized training.In this study,we proposed a novel multimodal deep learning approach,based on the FAST,for assessing suspected stroke patients exhibiting symptoms such as limb weakness,facial paresis,and speech disorders in acute settings.We collected a dataset comprising videos and audio recordings of emergency room patients performing designated limb movements,facial expressions,and speech tests based on the FAST.We compared the constructed deep learning model,which was designed to process multi-modal datasets,with six prior models that achieved good action classification performance,including the I3D,SlowFast,X3D,TPN,TimeSformer,and MViT.We found that the findings of our deep learning model had a higher clinical value compared with the other approaches.Moreover,the multi-modal model outperformed its single-module variants,highlighting the benefit of utilizing multiple types of patient data,such as action videos and speech audio.These results indicate that a multi-modal deep learning model combined with the FAST could greatly improve the accuracy and sensitivity of early stroke identification of stroke,thus providing a practical and powerful tool for assessing stroke patients in an emergency clinical setting.展开更多
With the rapid development of artificial intelligence,the Internet of Things(IoT)can deploy various machine learning algorithms for network and application management.In the IoT environment,many sensors and devices ge...With the rapid development of artificial intelligence,the Internet of Things(IoT)can deploy various machine learning algorithms for network and application management.In the IoT environment,many sensors and devices generatemassive data,but data security and privacy protection have become a serious challenge.Federated learning(FL)can achieve many intelligent IoT applications by training models on local devices and allowing AI training on distributed IoT devices without data sharing.This review aims to deeply explore the combination of FL and the IoT,and analyze the application of federated learning in the IoT from the aspects of security and privacy protection.In this paper,we first describe the potential advantages of FL and the challenges faced by current IoT systems in the fields of network burden and privacy security.Next,we focus on exploring and analyzing the advantages of the combination of FL on the Internet,including privacy security,attack detection,efficient communication of the IoT,and enhanced learning quality.We also list various application scenarios of FL on the IoT.Finally,we propose several open research challenges and possible solutions.展开更多
In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data silos.Vertical Federated Learning(VFL)has emerged as a promising approach...In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data silos.Vertical Federated Learning(VFL)has emerged as a promising approach to facilitate such collaboration,allowing diverse entities to collectively enhance machine learning models without the need to share sensitive training data.However,existing works have highlighted VFL’s susceptibility to privacy inference attacks,where an honest but curious server could potentially reconstruct a client’s raw data from embeddings uploaded by the client.This vulnerability poses a significant threat to VFL-based intelligent railway transportation systems.In this paper,we introduce SensFL,a novel privacy-enhancing method to against privacy inference attacks in VFL.Specifically,SensFL integrates regularization of the sensitivity of embeddings to the original data into the model training process,effectively limiting the information contained in shared embeddings.By reducing the sensitivity of embeddings to the original data,SensFL can effectively resist reverse privacy attacks and prevent the reconstruction of the original data from the embeddings.Extensive experiments were conducted on four distinct datasets and three different models to demonstrate the efficacy of SensFL.Experiment results show that SensFL can effectively mitigate privacy inference attacks while maintaining the accuracy of the primary learning task.These results underscore SensFL’s potential to advance privacy protection technologies within VFL-based intelligent railway systems,addressing critical security concerns in collaborative learning environments.展开更多
Mental health is a significant issue worldwide,and the utilization of technology to assist mental health has seen a growing trend.This aims to alleviate the workload on healthcare professionals and aid individuals.Num...Mental health is a significant issue worldwide,and the utilization of technology to assist mental health has seen a growing trend.This aims to alleviate the workload on healthcare professionals and aid individuals.Numerous applications have been developed to support the challenges in intelligent healthcare systems.However,because mental health data is sensitive,privacy concerns have emerged.Federated learning has gotten some attention.This research reviews the studies on federated learning and mental health related to solving the issue of intelligent healthcare systems.It explores various dimensions of federated learning in mental health,such as datasets(their types and sources),applications categorized based on mental health symptoms,federated mental health frameworks,federated machine learning,federated deep learning,and the benefits of federated learning in mental health applications.This research conducts surveys to evaluate the current state of mental health applications,mainly focusing on the role of Federated Learning(FL)and related privacy and data security concerns.The survey provides valuable insights into how these applications are emerging and evolving,specifically emphasizing FL’s impact.展开更多
Two-dimensional endoscopic images are susceptible to interferences such as specular reflections and monotonous texture illumination,hindering accurate three-dimensional lesion reconstruction by surgical robots.This st...Two-dimensional endoscopic images are susceptible to interferences such as specular reflections and monotonous texture illumination,hindering accurate three-dimensional lesion reconstruction by surgical robots.This study proposes a novel end-to-end disparity estimation model to address these challenges.Our approach combines a Pseudo-Siamese neural network architecture with pyramid dilated convolutions,integrating multi-scale image information to enhance robustness against lighting interferences.This study introduces a Pseudo-Siamese structure-based disparity regression model that simplifies left-right image comparison,improving accuracy and efficiency.The model was evaluated using a dataset of stereo endoscopic videos captured by the Da Vinci surgical robot,comprising simulated silicone heart sequences and real heart video data.Experimental results demonstrate significant improvement in the network’s resistance to lighting interference without substantially increasing parameters.Moreover,the model exhibited faster convergence during training,contributing to overall performance enhancement.This study advances endoscopic image processing accuracy and has potential implications for surgical robot applications in complex environments.展开更多
针对无监督环境下传统网络异常诊断算法存在异常点定位和异常数据分类准确率低等不足,通过设计一种基于改进Q-learning算法的无线网络异常诊断方法:首先基于ADU(Asynchronous Data Unit异步数据单元)单元采集无线网络的数据流,并提取数...针对无监督环境下传统网络异常诊断算法存在异常点定位和异常数据分类准确率低等不足,通过设计一种基于改进Q-learning算法的无线网络异常诊断方法:首先基于ADU(Asynchronous Data Unit异步数据单元)单元采集无线网络的数据流,并提取数据包特征;然后构建Q-learning算法模型探索状态值和奖励值的平衡点,利用SA(Simulated Annealing模拟退火)算法从全局视角对下一时刻状态进行精确识别;最后确定训练样本的联合分布概率,提升输出值的逼近性能以达到平衡探索与代价之间的均衡。测试结果显示:改进Q-learning算法的网络异常定位准确率均值达99.4%,在不同类型网络异常的分类精度和分类效率等方面,也优于三种传统网络异常诊断方法。展开更多
Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay di...Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.展开更多
As AI systems scale, the limitations of cloud-based architectures, including latency, bandwidth, and privacy concerns, demand decentralized alternatives. Federated learning (FL) and Edge AI provide a paradigm shift by...As AI systems scale, the limitations of cloud-based architectures, including latency, bandwidth, and privacy concerns, demand decentralized alternatives. Federated learning (FL) and Edge AI provide a paradigm shift by combining privacy preserving training with efficient, on device computation. This paper introduces a cutting-edge FL-edge integration framework, achieving a 10% to 15% increase in model accuracy and reducing communication costs by 25% in heterogeneous environments. Blockchain based secure aggregation ensures robust and tamper-proof model updates, while exploratory quantum AI techniques enhance computational efficiency. By addressing key challenges such as device variability and non-IID data, this work sets the stage for the next generation of adaptive, privacy-first AI systems, with applications in IoT, healthcare, and autonomous systems.展开更多
The burgeoning market for lithium-ion batteries has stimulated a growing need for more reliable battery performance monitoring. Accurate state-of-health(SOH) estimation is critical for ensuring battery operational per...The burgeoning market for lithium-ion batteries has stimulated a growing need for more reliable battery performance monitoring. Accurate state-of-health(SOH) estimation is critical for ensuring battery operational performance. Despite numerous data-driven methods reported in existing research for battery SOH estimation, these methods often exhibit inconsistent performance across different application scenarios. To address this issue and overcome the performance limitations of individual data-driven models,integrating multiple models for SOH estimation has received considerable attention. Ensemble learning(EL) typically leverages the strengths of multiple base models to achieve more robust and accurate outputs. However, the lack of a clear review of current research hinders the further development of ensemble methods in SOH estimation. Therefore, this paper comprehensively reviews multi-model ensemble learning methods for battery SOH estimation. First, existing ensemble methods are systematically categorized into 6 classes based on their combination strategies. Different realizations and underlying connections are meticulously analyzed for each category of EL methods, highlighting distinctions, innovations, and typical applications. Subsequently, these ensemble methods are comprehensively compared in terms of base models, combination strategies, and publication trends. Evaluations across 6 dimensions underscore the outstanding performance of stacking-based ensemble methods. Following this, these ensemble methods are further inspected from the perspectives of weighted ensemble and diversity, aiming to inspire potential approaches for enhancing ensemble performance. Moreover, addressing challenges such as base model selection, measuring model robustness and uncertainty, and interpretability of ensemble models in practical applications is emphasized. Finally, future research prospects are outlined, specifically noting that deep learning ensemble is poised to advance ensemble methods for battery SOH estimation. The convergence of advanced machine learning with ensemble learning is anticipated to yield valuable avenues for research. Accelerated research in ensemble learning holds promising prospects for achieving more accurate and reliable battery SOH estimation under real-world conditions.展开更多
In computer vision and artificial intelligence,automatic facial expression-based emotion identification of humans has become a popular research and industry problem.Recent demonstrations and applications in several fi...In computer vision and artificial intelligence,automatic facial expression-based emotion identification of humans has become a popular research and industry problem.Recent demonstrations and applications in several fields,including computer games,smart homes,expression analysis,gesture recognition,surveillance films,depression therapy,patientmonitoring,anxiety,and others,have brought attention to its significant academic and commercial importance.This study emphasizes research that has only employed facial images for face expression recognition(FER),because facial expressions are a basic way that people communicate meaning to each other.The immense achievement of deep learning has resulted in a growing use of its much architecture to enhance efficiency.This review is on machine learning,deep learning,and hybrid methods’use of preprocessing,augmentation techniques,and feature extraction for temporal properties of successive frames of data.The following section gives a brief summary of assessment criteria that are accessible to the public and then compares them with benchmark results the most trustworthy way to assess FER-related research topics statistically.In this review,a brief synopsis of the subject matter may be beneficial for novices in the field of FER as well as seasoned scholars seeking fruitful avenues for further investigation.The information conveys fundamental knowledge and provides a comprehensive understanding of the most recent state-of-the-art research.展开更多
The current deep learning models for braced excavation cannot predict deformation from the beginning of excavation due to the need for a substantial corpus of sufficient historical data for training purposes.To addres...The current deep learning models for braced excavation cannot predict deformation from the beginning of excavation due to the need for a substantial corpus of sufficient historical data for training purposes.To address this issue,this study proposes a transfer learning model based on a sequence-to-sequence twodimensional(2D)convolutional long short-term memory neural network(S2SCL2D).The model can use the existing data from other adjacent similar excavations to achieve wall deflection prediction once a limited amount of monitoring data from the target excavation has been recorded.In the absence of adjacent excavation data,numerical simulation data from the target project can be employed instead.A weight update strategy is proposed to improve the prediction accuracy by integrating the stochastic gradient masking with an early stopping mechanism.To illustrate the proposed methodology,an excavation project in Hangzhou,China is adopted.The proposed deep transfer learning model,which uses either adjacent excavation data or numerical simulation data as the source domain,shows a significant improvement in performance when compared to the non-transfer learning model.Using the simulation data from the target project even leads to better prediction performance than using the actual monitoring data from other adjacent excavations.The results demonstrate that the proposed model can reasonably predict the deformation with limited data from the target project.展开更多
As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigat...As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigation of degradation mechanisms.However,dynamic operating conditions,cell-to-cell inconsistencies,and limited availability of labeled data have posed significant challenges to accurate and robust prognosis and diagnosis.Herein,we introduce a time-series-decomposition-based ensembled lightweight learning model(TELL-Me),which employs a synergistic dual-module framework to facilitate accurate and reliable forecasting.The feature module formulates features with physical implications and sheds light on battery aging mechanisms,while the gradient module monitors capacity degradation rates and captures aging trend.TELL-Me achieves high accuracy in end-of-life prediction using minimal historical data from a single battery without requiring offline training dataset,and demonstrates impressive generality and robustness across various operating conditions and battery types.Additionally,by correlating feature contributions with degradation mechanisms across different datasets,TELL-Me is endowed with the diagnostic ability that not only enhances prediction reliability but also provides critical insights into the design and optimization of next-generation batteries.展开更多
Accurate channel state information(CSI)is crucial for 6G wireless communication systems to accommodate the growing demands of mobile broadband services.In massive multiple-input multiple-output(MIMO)systems,traditiona...Accurate channel state information(CSI)is crucial for 6G wireless communication systems to accommodate the growing demands of mobile broadband services.In massive multiple-input multiple-output(MIMO)systems,traditional CSI feedback approaches face challenges such as performance degradation due to feedback delay and channel aging caused by user mobility.To address these issues,we propose a novel spatio-temporal predictive network(STPNet)that jointly integrates CSI feedback and prediction modules.STPNet employs stacked Inception modules to learn the spatial correlation and temporal evolution of CSI,which captures both the local and the global spatiotemporal features.In addition,the signal-to-noise ratio(SNR)adaptive module is designed to adapt flexibly to diverse feedback channel conditions.Simulation results demonstrate that STPNet outperforms existing channel prediction methods under various channel conditions.展开更多
Federated learning is a machine learning framework designed to protect privacy by keeping training data on clients’devices without sharing private data.It trains a global model through collaboration between clients a...Federated learning is a machine learning framework designed to protect privacy by keeping training data on clients’devices without sharing private data.It trains a global model through collaboration between clients and the server.However,the presence of data heterogeneity can lead to inefficient model training and even reduce the final model’s accuracy and generalization capability.Meanwhile,data scarcity can result in suboptimal cluster distributions for few-shot clients in centralized clustering tasks,and standalone personalization tasks may cause severe overfitting issues.To address these limitations,we introduce a federated learning dual optimization model based on clustering and personalization strategy(FedCPS).FedCPS adopts a decentralized approach,where clients identify their cluster membership locally without relying on a centralized clustering algorithm.Building on this,FedCPS introduces personalized training tasks locally,adding a regularization term to control deviations between local and cluster models.This improves the generalization ability of the final model while mitigating overfitting.The use of weight-sharing techniques also reduces the computational cost of central machines.Experimental results on MNIST,FMNIST,CIFAR10,and CIFAR100 datasets demonstrate that our method achieves better personalization effects compared to other personalized federated learning methods,with an average test accuracy improvement of 0.81%–2.96%.Meanwhile,we adjusted the proportion of few-shot clients to evaluate the impact on accuracy across different methods.The experiments show that FedCPS reduces accuracy by only 0.2%–3.7%,compared to 2.1%–10%for existing methods.Our method demonstrates its advantages across diverse data environments.展开更多
Arsenic(As)pollution in soils is a pervasive environmental issue.Biochar immobilization offers a promising solution for addressing soil As contamination.The efficiency of biochar in immobilizing As in soils primarily ...Arsenic(As)pollution in soils is a pervasive environmental issue.Biochar immobilization offers a promising solution for addressing soil As contamination.The efficiency of biochar in immobilizing As in soils primarily hinges on the characteristics of both the soil and the biochar.However,the influence of a specific property on As immobilization varies among different studies,and the development and application of arsenic passivation materials based on biochar often rely on empirical knowledge.To enhance immobilization efficiency and reduce labor and time costs,a machine learning(ML)model was employed to predict As immobilization efficiency before biochar application.In this study,we collected a dataset comprising 182 data points on As immobilization efficiency from 17 publications to construct three ML models.The results demonstrated that the random forest(RF)model outperformed gradient boost regression tree and support vector regression models in predictive performance.Relative importance analysis and partial dependence plots based on the RF model were conducted to identify the most crucial factors influencing As immobilization.These findings highlighted the significant roles of biochar application time and biochar pH in As immobilization efficiency in soils.Furthermore,the study revealed that Fe-modified biochar exhibited a substantial improvement in As immobilization.These insights can facilitate targeted biochar property design and optimization of biochar application conditions to enhance As immobilization efficiency.展开更多
Artificial neural networks are capable of machine learning by simulating the hiera rchical structure of the human brain.To enable learning by brain and machine,it is essential to accurately identify and correct the pr...Artificial neural networks are capable of machine learning by simulating the hiera rchical structure of the human brain.To enable learning by brain and machine,it is essential to accurately identify and correct the prediction errors,referred to as credit assignment(Lillicrap et al.,2020).It is critical to develop artificial intelligence by understanding how the brain deals with credit assignment in neuroscience.展开更多
Liposomes serve as critical carriers for drugs and vaccines,with their biological effects influenced by their size.The microfluidic method,renowned for its precise control,reproducibility,and scalability,has been wide...Liposomes serve as critical carriers for drugs and vaccines,with their biological effects influenced by their size.The microfluidic method,renowned for its precise control,reproducibility,and scalability,has been widely employed for liposome preparation.Although some studies have explored factors affecting liposomal size in microfluidic processes,most focus on small-sized liposomes,predominantly through experimental data analysis.However,the production of larger liposomes,which are equally significant,remains underexplored.In this work,we thoroughly investigate multiple variables influencing liposome size during microfluidic preparation and develop a machine learning(ML)model capable of accurately predicting liposomal size.Experimental validation was conducted using a staggered herringbone micromixer(SHM)chip.Our findings reveal that most investigated variables significantly influence liposomal size,often interrelating in complex ways.We evaluated the predictive performance of several widely-used ML algorithms,including ensemble methods,through cross-validation(CV)for both lipo-some size and polydispersity index(PDI).A standalone dataset was experimentally validated to assess the accuracy of the ML predictions,with results indicating that ensemble algorithms provided the most reliable predictions.Specifically,gradient boosting was selected for size prediction,while random forest was employed for PDI prediction.We successfully produced uniform large(600 nm)and small(100 nm)liposomes using the optimised experimental conditions derived from the ML models.In conclusion,this study presents a robust methodology that enables precise control over liposome size distribution,of-fering valuable insights for medicinal research applications.展开更多
Excellent detonation performances and low sensitivity are prerequisites for the deployment of energetic materials.Exploring the underlying factors that affect impact sensitivity and detonation performances as well as ...Excellent detonation performances and low sensitivity are prerequisites for the deployment of energetic materials.Exploring the underlying factors that affect impact sensitivity and detonation performances as well as exploring how to obtain materials with desired properties remains a long-term challenge.Machine learning with its ability to solve complex tasks and perform robust data processing can reveal the relationship between performance and descriptive indicators,potentially accelerating the development process of energetic materials.In this background,impact sensitivity,detonation performances,and 28 physicochemical parameters for 222 energetic materials from density functional theory calculations and published literature were sorted out.Four machine learning algorithms were employed to predict various properties of energetic materials,including impact sensitivity,detonation velocity,detonation pressure,and Gurney energy.Analysis of Pearson coefficients and feature importance showed that the heat of explosion,oxygen balance,decomposition products,and HOMO energy levels have a strong correlation with the impact sensitivity of energetic materials.Oxygen balance,decomposition products,and density have a strong correlation with detonation performances.Utilizing impact sensitivity of 2,3,4-trinitrotoluene and the detonation performances of 2,4,6-trinitrobenzene-1,3,5-triamine as the benchmark,the analysis of feature importance rankings and statistical data revealed the optimal range of key features balancing impact sensitivity and detonation performances:oxygen balance values should be between-40%and-30%,density should range from 1.66 to 1.72 g/cm^(3),HOMO energy levels should be between-6.34 and-6.31 eV,and lipophilicity should be between-1.0 and 0.1,4.49 and 5.59.These findings not only offer important insights into the impact sensitivity and detonation performances of energetic materials,but also provide a theoretical guidance paradigm for the design and development of new energetic materials with optimal detonation performances and reduced sensitivity.展开更多
基金supported by the Ministry of Science and Technology of China,No.2020AAA0109605(to XL)Meizhou Major Scientific and Technological Innovation PlatformsProjects of Guangdong Provincial Science & Technology Plan Projects,No.2019A0102005(to HW).
文摘Early identification and treatment of stroke can greatly improve patient outcomes and quality of life.Although clinical tests such as the Cincinnati Pre-hospital Stroke Scale(CPSS)and the Face Arm Speech Test(FAST)are commonly used for stroke screening,accurate administration is dependent on specialized training.In this study,we proposed a novel multimodal deep learning approach,based on the FAST,for assessing suspected stroke patients exhibiting symptoms such as limb weakness,facial paresis,and speech disorders in acute settings.We collected a dataset comprising videos and audio recordings of emergency room patients performing designated limb movements,facial expressions,and speech tests based on the FAST.We compared the constructed deep learning model,which was designed to process multi-modal datasets,with six prior models that achieved good action classification performance,including the I3D,SlowFast,X3D,TPN,TimeSformer,and MViT.We found that the findings of our deep learning model had a higher clinical value compared with the other approaches.Moreover,the multi-modal model outperformed its single-module variants,highlighting the benefit of utilizing multiple types of patient data,such as action videos and speech audio.These results indicate that a multi-modal deep learning model combined with the FAST could greatly improve the accuracy and sensitivity of early stroke identification of stroke,thus providing a practical and powerful tool for assessing stroke patients in an emergency clinical setting.
基金supported by the Shandong Province Science and Technology Project(2023TSGC0509,2022TSGC2234)Qingdao Science and Technology Plan Project(23-1-5-yqpy-2-qy)Open Topic Grants of Anhui Province Key Laboratory of Intelligent Building&Building Energy Saving,Anhui Jianzhu University(IBES2024KF08).
文摘With the rapid development of artificial intelligence,the Internet of Things(IoT)can deploy various machine learning algorithms for network and application management.In the IoT environment,many sensors and devices generatemassive data,but data security and privacy protection have become a serious challenge.Federated learning(FL)can achieve many intelligent IoT applications by training models on local devices and allowing AI training on distributed IoT devices without data sharing.This review aims to deeply explore the combination of FL and the IoT,and analyze the application of federated learning in the IoT from the aspects of security and privacy protection.In this paper,we first describe the potential advantages of FL and the challenges faced by current IoT systems in the fields of network burden and privacy security.Next,we focus on exploring and analyzing the advantages of the combination of FL on the Internet,including privacy security,attack detection,efficient communication of the IoT,and enhanced learning quality.We also list various application scenarios of FL on the IoT.Finally,we propose several open research challenges and possible solutions.
基金supported by Systematic Major Project of Shuohuang Railway Development Co.,Ltd.,National Energy Group(Grant Number:SHTL-23-31)Beijing Natural Science Foundation(U22B2027).
文摘In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data silos.Vertical Federated Learning(VFL)has emerged as a promising approach to facilitate such collaboration,allowing diverse entities to collectively enhance machine learning models without the need to share sensitive training data.However,existing works have highlighted VFL’s susceptibility to privacy inference attacks,where an honest but curious server could potentially reconstruct a client’s raw data from embeddings uploaded by the client.This vulnerability poses a significant threat to VFL-based intelligent railway transportation systems.In this paper,we introduce SensFL,a novel privacy-enhancing method to against privacy inference attacks in VFL.Specifically,SensFL integrates regularization of the sensitivity of embeddings to the original data into the model training process,effectively limiting the information contained in shared embeddings.By reducing the sensitivity of embeddings to the original data,SensFL can effectively resist reverse privacy attacks and prevent the reconstruction of the original data from the embeddings.Extensive experiments were conducted on four distinct datasets and three different models to demonstrate the efficacy of SensFL.Experiment results show that SensFL can effectively mitigate privacy inference attacks while maintaining the accuracy of the primary learning task.These results underscore SensFL’s potential to advance privacy protection technologies within VFL-based intelligent railway systems,addressing critical security concerns in collaborative learning environments.
文摘Mental health is a significant issue worldwide,and the utilization of technology to assist mental health has seen a growing trend.This aims to alleviate the workload on healthcare professionals and aid individuals.Numerous applications have been developed to support the challenges in intelligent healthcare systems.However,because mental health data is sensitive,privacy concerns have emerged.Federated learning has gotten some attention.This research reviews the studies on federated learning and mental health related to solving the issue of intelligent healthcare systems.It explores various dimensions of federated learning in mental health,such as datasets(their types and sources),applications categorized based on mental health symptoms,federated mental health frameworks,federated machine learning,federated deep learning,and the benefits of federated learning in mental health applications.This research conducts surveys to evaluate the current state of mental health applications,mainly focusing on the role of Federated Learning(FL)and related privacy and data security concerns.The survey provides valuable insights into how these applications are emerging and evolving,specifically emphasizing FL’s impact.
基金Supported by Sichuan Science and Technology Program(2023YFSY0026,2023YFH0004)Supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korean government(MSIT)(No.RS-2022-00155885,Artificial Intelligence Convergence Innovation Human Resources Development(Hanyang University ERICA)).
文摘Two-dimensional endoscopic images are susceptible to interferences such as specular reflections and monotonous texture illumination,hindering accurate three-dimensional lesion reconstruction by surgical robots.This study proposes a novel end-to-end disparity estimation model to address these challenges.Our approach combines a Pseudo-Siamese neural network architecture with pyramid dilated convolutions,integrating multi-scale image information to enhance robustness against lighting interferences.This study introduces a Pseudo-Siamese structure-based disparity regression model that simplifies left-right image comparison,improving accuracy and efficiency.The model was evaluated using a dataset of stereo endoscopic videos captured by the Da Vinci surgical robot,comprising simulated silicone heart sequences and real heart video data.Experimental results demonstrate significant improvement in the network’s resistance to lighting interference without substantially increasing parameters.Moreover,the model exhibited faster convergence during training,contributing to overall performance enhancement.This study advances endoscopic image processing accuracy and has potential implications for surgical robot applications in complex environments.
文摘针对无监督环境下传统网络异常诊断算法存在异常点定位和异常数据分类准确率低等不足,通过设计一种基于改进Q-learning算法的无线网络异常诊断方法:首先基于ADU(Asynchronous Data Unit异步数据单元)单元采集无线网络的数据流,并提取数据包特征;然后构建Q-learning算法模型探索状态值和奖励值的平衡点,利用SA(Simulated Annealing模拟退火)算法从全局视角对下一时刻状态进行精确识别;最后确定训练样本的联合分布概率,提升输出值的逼近性能以达到平衡探索与代价之间的均衡。测试结果显示:改进Q-learning算法的网络异常定位准确率均值达99.4%,在不同类型网络异常的分类精度和分类效率等方面,也优于三种传统网络异常诊断方法。
文摘Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.
文摘As AI systems scale, the limitations of cloud-based architectures, including latency, bandwidth, and privacy concerns, demand decentralized alternatives. Federated learning (FL) and Edge AI provide a paradigm shift by combining privacy preserving training with efficient, on device computation. This paper introduces a cutting-edge FL-edge integration framework, achieving a 10% to 15% increase in model accuracy and reducing communication costs by 25% in heterogeneous environments. Blockchain based secure aggregation ensures robust and tamper-proof model updates, while exploratory quantum AI techniques enhance computational efficiency. By addressing key challenges such as device variability and non-IID data, this work sets the stage for the next generation of adaptive, privacy-first AI systems, with applications in IoT, healthcare, and autonomous systems.
基金National Natural Science Foundation of China (52075420)Fundamental Research Funds for the Central Universities (xzy022023049)National Key Research and Development Program of China (2023YFB3408600)。
文摘The burgeoning market for lithium-ion batteries has stimulated a growing need for more reliable battery performance monitoring. Accurate state-of-health(SOH) estimation is critical for ensuring battery operational performance. Despite numerous data-driven methods reported in existing research for battery SOH estimation, these methods often exhibit inconsistent performance across different application scenarios. To address this issue and overcome the performance limitations of individual data-driven models,integrating multiple models for SOH estimation has received considerable attention. Ensemble learning(EL) typically leverages the strengths of multiple base models to achieve more robust and accurate outputs. However, the lack of a clear review of current research hinders the further development of ensemble methods in SOH estimation. Therefore, this paper comprehensively reviews multi-model ensemble learning methods for battery SOH estimation. First, existing ensemble methods are systematically categorized into 6 classes based on their combination strategies. Different realizations and underlying connections are meticulously analyzed for each category of EL methods, highlighting distinctions, innovations, and typical applications. Subsequently, these ensemble methods are comprehensively compared in terms of base models, combination strategies, and publication trends. Evaluations across 6 dimensions underscore the outstanding performance of stacking-based ensemble methods. Following this, these ensemble methods are further inspected from the perspectives of weighted ensemble and diversity, aiming to inspire potential approaches for enhancing ensemble performance. Moreover, addressing challenges such as base model selection, measuring model robustness and uncertainty, and interpretability of ensemble models in practical applications is emphasized. Finally, future research prospects are outlined, specifically noting that deep learning ensemble is poised to advance ensemble methods for battery SOH estimation. The convergence of advanced machine learning with ensemble learning is anticipated to yield valuable avenues for research. Accelerated research in ensemble learning holds promising prospects for achieving more accurate and reliable battery SOH estimation under real-world conditions.
文摘In computer vision and artificial intelligence,automatic facial expression-based emotion identification of humans has become a popular research and industry problem.Recent demonstrations and applications in several fields,including computer games,smart homes,expression analysis,gesture recognition,surveillance films,depression therapy,patientmonitoring,anxiety,and others,have brought attention to its significant academic and commercial importance.This study emphasizes research that has only employed facial images for face expression recognition(FER),because facial expressions are a basic way that people communicate meaning to each other.The immense achievement of deep learning has resulted in a growing use of its much architecture to enhance efficiency.This review is on machine learning,deep learning,and hybrid methods’use of preprocessing,augmentation techniques,and feature extraction for temporal properties of successive frames of data.The following section gives a brief summary of assessment criteria that are accessible to the public and then compares them with benchmark results the most trustworthy way to assess FER-related research topics statistically.In this review,a brief synopsis of the subject matter may be beneficial for novices in the field of FER as well as seasoned scholars seeking fruitful avenues for further investigation.The information conveys fundamental knowledge and provides a comprehensive understanding of the most recent state-of-the-art research.
基金supported by the National Key Research and Development Program of China(Grant No.2023YFC3009400)the National Natural Science Foundation of China(Grant Nos.42307218 and U2239251).
文摘The current deep learning models for braced excavation cannot predict deformation from the beginning of excavation due to the need for a substantial corpus of sufficient historical data for training purposes.To address this issue,this study proposes a transfer learning model based on a sequence-to-sequence twodimensional(2D)convolutional long short-term memory neural network(S2SCL2D).The model can use the existing data from other adjacent similar excavations to achieve wall deflection prediction once a limited amount of monitoring data from the target excavation has been recorded.In the absence of adjacent excavation data,numerical simulation data from the target project can be employed instead.A weight update strategy is proposed to improve the prediction accuracy by integrating the stochastic gradient masking with an early stopping mechanism.To illustrate the proposed methodology,an excavation project in Hangzhou,China is adopted.The proposed deep transfer learning model,which uses either adjacent excavation data or numerical simulation data as the source domain,shows a significant improvement in performance when compared to the non-transfer learning model.Using the simulation data from the target project even leads to better prediction performance than using the actual monitoring data from other adjacent excavations.The results demonstrate that the proposed model can reasonably predict the deformation with limited data from the target project.
基金supported by the National Natural Science Foundation of China(22379021 and 22479021)。
文摘As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigation of degradation mechanisms.However,dynamic operating conditions,cell-to-cell inconsistencies,and limited availability of labeled data have posed significant challenges to accurate and robust prognosis and diagnosis.Herein,we introduce a time-series-decomposition-based ensembled lightweight learning model(TELL-Me),which employs a synergistic dual-module framework to facilitate accurate and reliable forecasting.The feature module formulates features with physical implications and sheds light on battery aging mechanisms,while the gradient module monitors capacity degradation rates and captures aging trend.TELL-Me achieves high accuracy in end-of-life prediction using minimal historical data from a single battery without requiring offline training dataset,and demonstrates impressive generality and robustness across various operating conditions and battery types.Additionally,by correlating feature contributions with degradation mechanisms across different datasets,TELL-Me is endowed with the diagnostic ability that not only enhances prediction reliability but also provides critical insights into the design and optimization of next-generation batteries.
基金supported in part by the Natural Science Foundation of China under Grant Nos.U2468201 and 62221001ZTE Industry-University-Institute Cooperation Funds under Grant No.IA20240420002。
文摘Accurate channel state information(CSI)is crucial for 6G wireless communication systems to accommodate the growing demands of mobile broadband services.In massive multiple-input multiple-output(MIMO)systems,traditional CSI feedback approaches face challenges such as performance degradation due to feedback delay and channel aging caused by user mobility.To address these issues,we propose a novel spatio-temporal predictive network(STPNet)that jointly integrates CSI feedback and prediction modules.STPNet employs stacked Inception modules to learn the spatial correlation and temporal evolution of CSI,which captures both the local and the global spatiotemporal features.In addition,the signal-to-noise ratio(SNR)adaptive module is designed to adapt flexibly to diverse feedback channel conditions.Simulation results demonstrate that STPNet outperforms existing channel prediction methods under various channel conditions.
基金supported by the Foundation of President of Hebei University(XZJJ202303).
文摘Federated learning is a machine learning framework designed to protect privacy by keeping training data on clients’devices without sharing private data.It trains a global model through collaboration between clients and the server.However,the presence of data heterogeneity can lead to inefficient model training and even reduce the final model’s accuracy and generalization capability.Meanwhile,data scarcity can result in suboptimal cluster distributions for few-shot clients in centralized clustering tasks,and standalone personalization tasks may cause severe overfitting issues.To address these limitations,we introduce a federated learning dual optimization model based on clustering and personalization strategy(FedCPS).FedCPS adopts a decentralized approach,where clients identify their cluster membership locally without relying on a centralized clustering algorithm.Building on this,FedCPS introduces personalized training tasks locally,adding a regularization term to control deviations between local and cluster models.This improves the generalization ability of the final model while mitigating overfitting.The use of weight-sharing techniques also reduces the computational cost of central machines.Experimental results on MNIST,FMNIST,CIFAR10,and CIFAR100 datasets demonstrate that our method achieves better personalization effects compared to other personalized federated learning methods,with an average test accuracy improvement of 0.81%–2.96%.Meanwhile,we adjusted the proportion of few-shot clients to evaluate the impact on accuracy across different methods.The experiments show that FedCPS reduces accuracy by only 0.2%–3.7%,compared to 2.1%–10%for existing methods.Our method demonstrates its advantages across diverse data environments.
基金supported by the National Key Research and Development Program of China(No.2020YFC1808701).
文摘Arsenic(As)pollution in soils is a pervasive environmental issue.Biochar immobilization offers a promising solution for addressing soil As contamination.The efficiency of biochar in immobilizing As in soils primarily hinges on the characteristics of both the soil and the biochar.However,the influence of a specific property on As immobilization varies among different studies,and the development and application of arsenic passivation materials based on biochar often rely on empirical knowledge.To enhance immobilization efficiency and reduce labor and time costs,a machine learning(ML)model was employed to predict As immobilization efficiency before biochar application.In this study,we collected a dataset comprising 182 data points on As immobilization efficiency from 17 publications to construct three ML models.The results demonstrated that the random forest(RF)model outperformed gradient boost regression tree and support vector regression models in predictive performance.Relative importance analysis and partial dependence plots based on the RF model were conducted to identify the most crucial factors influencing As immobilization.These findings highlighted the significant roles of biochar application time and biochar pH in As immobilization efficiency in soils.Furthermore,the study revealed that Fe-modified biochar exhibited a substantial improvement in As immobilization.These insights can facilitate targeted biochar property design and optimization of biochar application conditions to enhance As immobilization efficiency.
基金supported by the National Natural Science Foundation of China,No.62276089。
文摘Artificial neural networks are capable of machine learning by simulating the hiera rchical structure of the human brain.To enable learning by brain and machine,it is essential to accurately identify and correct the prediction errors,referred to as credit assignment(Lillicrap et al.,2020).It is critical to develop artificial intelligence by understanding how the brain deals with credit assignment in neuroscience.
基金supported by the National Key Research and Development Plan of the Ministry of Science and Technology,China(Grant No.:2022YFE0125300)the National Natural Science Foundation of China(Grant No:81690262)+2 种基金the National Science and Technology Major Project,China(Grant No.:2017ZX09201004-021)the Open Project of National facility for Translational Medicine(Shanghai),China(Grant No.:TMSK-2021-104)Shanghai Jiao Tong University STAR Grant,China(Grant Nos.:YG2022ZD024 and YG2022QN111).
文摘Liposomes serve as critical carriers for drugs and vaccines,with their biological effects influenced by their size.The microfluidic method,renowned for its precise control,reproducibility,and scalability,has been widely employed for liposome preparation.Although some studies have explored factors affecting liposomal size in microfluidic processes,most focus on small-sized liposomes,predominantly through experimental data analysis.However,the production of larger liposomes,which are equally significant,remains underexplored.In this work,we thoroughly investigate multiple variables influencing liposome size during microfluidic preparation and develop a machine learning(ML)model capable of accurately predicting liposomal size.Experimental validation was conducted using a staggered herringbone micromixer(SHM)chip.Our findings reveal that most investigated variables significantly influence liposomal size,often interrelating in complex ways.We evaluated the predictive performance of several widely-used ML algorithms,including ensemble methods,through cross-validation(CV)for both lipo-some size and polydispersity index(PDI).A standalone dataset was experimentally validated to assess the accuracy of the ML predictions,with results indicating that ensemble algorithms provided the most reliable predictions.Specifically,gradient boosting was selected for size prediction,while random forest was employed for PDI prediction.We successfully produced uniform large(600 nm)and small(100 nm)liposomes using the optimised experimental conditions derived from the ML models.In conclusion,this study presents a robust methodology that enables precise control over liposome size distribution,of-fering valuable insights for medicinal research applications.
基金supported by the Fundamental Research Funds for the Central Universities(Grant No.2682024GF019)。
文摘Excellent detonation performances and low sensitivity are prerequisites for the deployment of energetic materials.Exploring the underlying factors that affect impact sensitivity and detonation performances as well as exploring how to obtain materials with desired properties remains a long-term challenge.Machine learning with its ability to solve complex tasks and perform robust data processing can reveal the relationship between performance and descriptive indicators,potentially accelerating the development process of energetic materials.In this background,impact sensitivity,detonation performances,and 28 physicochemical parameters for 222 energetic materials from density functional theory calculations and published literature were sorted out.Four machine learning algorithms were employed to predict various properties of energetic materials,including impact sensitivity,detonation velocity,detonation pressure,and Gurney energy.Analysis of Pearson coefficients and feature importance showed that the heat of explosion,oxygen balance,decomposition products,and HOMO energy levels have a strong correlation with the impact sensitivity of energetic materials.Oxygen balance,decomposition products,and density have a strong correlation with detonation performances.Utilizing impact sensitivity of 2,3,4-trinitrotoluene and the detonation performances of 2,4,6-trinitrobenzene-1,3,5-triamine as the benchmark,the analysis of feature importance rankings and statistical data revealed the optimal range of key features balancing impact sensitivity and detonation performances:oxygen balance values should be between-40%and-30%,density should range from 1.66 to 1.72 g/cm^(3),HOMO energy levels should be between-6.34 and-6.31 eV,and lipophilicity should be between-1.0 and 0.1,4.49 and 5.59.These findings not only offer important insights into the impact sensitivity and detonation performances of energetic materials,but also provide a theoretical guidance paradigm for the design and development of new energetic materials with optimal detonation performances and reduced sensitivity.
