Objective:To assess aptamer-based assays for diagnosing latent tuberculosis infection(LTBI).Methods:Literature from Medline,ScienceDirect,and Scopus,covering publications from January 1,2012,to December 31,2023,was ex...Objective:To assess aptamer-based assays for diagnosing latent tuberculosis infection(LTBI).Methods:Literature from Medline,ScienceDirect,and Scopus,covering publications from January 1,2012,to December 31,2023,was examined.This review evaluates different aptamers,biomarkers,sample types,sample sizes,reference assays,and the assays'sensitivity and specificity.By using the Quality Assessment of Diagnostic Accuracy Studies 2,the risk of bias in each study was evaluated.Results:Aptamer-based assays generally showed a sensitivity of 90%(95%CI:75%-100%)and specificity of 90%(95%CI:50%-100%),where optical aptasensor showed the highest sensitivity and specificity at 100%.Serum samples were frequently used to enhance antigen detectability,improving the assay’s performance.Meanwhile,HspX was the most studied biomarker,followed by MPT64,and IFN-γ.Conclusions:Aptamer-based assays could be reliable alternatives to current LTBI detection methods,but further research is needed to validate their clinical efficacy.展开更多
Objectives:The study aimed to explore social alienation types in patients with inflammatorybowel diseases(IBD)and identify influencingfactors.Methods:This cross-sectional study was conducted using purposive sampling a...Objectives:The study aimed to explore social alienation types in patients with inflammatorybowel diseases(IBD)and identify influencingfactors.Methods:This cross-sectional study was conducted using purposive sampling among patients with IBD from July 2022 to July 2023.Patients were assessed using the Generalized Social Alienation Scale(GSAS),the Brief Illness Perception Questionnaire(B-IPQ),the Hospital Anxiety and Depression Scale(HADS),and the Medical Coping Modes Questionnaire(MCMQ).Demographic and disease-related characteristics were also collected.Latent profileanalysis(LPA)was used to identify potential subgroups of social alienation.Univariate analysis and multicollinearity analysis were conducted to explore the influencing factors,followed by multiple regression analysis to evaluate the effect of influencingfactors on social alienation.Results:Three distinct profilesof social alienation were identified:integrated-low alienation group(n=61,20.20%),accommodative-moderate alienation group(n=195,64.57%),and maladaptive-high alienation group(n=46,15.23%).Seven characteristics were associated with the profile’smembership:self-perceived financialstress,malnutrition risk,disease duration,illness comprehensibility,anxiety,depression,and acceptance-resignation coping mode.Conclusions:Patients with IBD were categorized into three subgroups based on social alienation levels.Financial stress,malnutrition risk,disease duration,illness comprehensibility,anxiety,depression,and acceptance-resignation coping mode were key predictors of the subgroup membership.Targeted interventions should be developed to mitigate the negative effects of social alienation,with a focus on improving illness perception,alleviating anxiety and depression,and promoting effective coping strategies.展开更多
In the field of organic synthesis,the core objective of retrosynthetic methods is to deduce possible synthetic routes and precursor molecules for complex target molecules.Traditional retrosynthetic methods,such as tem...In the field of organic synthesis,the core objective of retrosynthetic methods is to deduce possible synthetic routes and precursor molecules for complex target molecules.Traditional retrosynthetic methods,such as template-based retrosynthesis,have high accuracy and interpretability in specific types of reactions but are limited by the scope of the template library,making it difficult to adapt to new or uncommon reaction types.Moreover,sequence-to-sequence retrosynthetic prediction methods,although they enhance the flexibility of prediction,often overlook the complexity of molecular graph structures and the actual interactions between atoms,which limits the accuracy and reliability of the predictions.To address these limitations,this paper proposes a Molecular Retrosynthesis Top-k Prediction based on the Latent Generation Process(MRLGP)that uses latent variables from graph neural networks to model the generation process and produce diverse set of reactants.Utilising an encoding method based on Graphormer,the authors have also introduced topology-aware positional encoding to better capture the interactions between atomic nodes in the molecular graph structure,thereby more accurately simulating the retrosynthetic process.The MRLGP model significantly enhances the accuracy and diversity of predictions by correlating discrete latent variables with the reactant generation process and progressively constructing molecular graphs using a variational autoregressive decoder.Experimental results on benchmark datasets such as USPTO-50k,USPTO-Full,and USPTO-DIVERSE demonstrate that MRLGP outperforms baseline models on multiple Top-k evaluation metrics.Additionally,ablation experiments conducted on the USPTO-50K dataset further validate the effectiveness of the methods used in the encoder and decoder parts of the model.展开更多
Objectives:Adolescents’aggression is widely studied,the underlying heterogeneity of aggression among rural Chinese boarding students remains unexplored.This study investigates the latent profiles of Chinese rural boa...Objectives:Adolescents’aggression is widely studied,the underlying heterogeneity of aggression among rural Chinese boarding students remains unexplored.This study investigates the latent profiles of Chinese rural boarding junior high school students’aggression and its correlations with moral disengagement and preference for solitude.Methods:A cross-sectional survey was conducted from 04–22 April 2022,using a convenient sampling method among 516 junior high school students from six Chinese rural boarding schools.The survey included the Aggression Questionnaire,the Moral Disengagement Scale(MDS),and the Preference for Solitude Scale(PSS).Results:Participants were divided into three latent profiles:low(36.0%),medium(50.9%),and high aggression levels(13.1%).Compared with low aggression,students who felt left-behind(minors who stay in the rural areas while one or both parents migrated to the urban areas for the work)accounted for a larger proportion in the medium aggression profile.The higher the grade level and the lower the educational level of the students’parents,the greater proportion of students in the medium and high aggression profiles.Additionally,students with high moral disengagement and preference for solitude showed a significant association with the medium aggression and high aggression profiles.Conclusions:The results demonstrate the significant group heterogeneity of aggression groups in Chinese rural boarding junior high school students.Targeted prevention and intervention measures can be carried out according to feeling left-behind,grade level,parents’education,and MDS and PSS scores.展开更多
Background:This study addresses the pressing need to understand the nuanced relationship between‘mattering’—the perception of being significant to others—and problematic internet use(PIU)among university students....Background:This study addresses the pressing need to understand the nuanced relationship between‘mattering’—the perception of being significant to others—and problematic internet use(PIU)among university students.Unlike previous research that has primarily employed variable-centered approaches,this study first adopts a person-centered approach using Latent Profile Analysis(LPA)to identify distinct mattering profiles.Subsequently,through variable-centered analyses,these profiles are examined in relation to different types of PIU—specifically problematic social media use(PSMU)and problematic gaming(PG)—as well as adaptability.Methods:Data were collected from 3587 university students across 19 universities in China.Participants completed three mattering-related scales(General Mattering Scale,Anti-Mattering Scale,and Fear of Not Mattering Inventory),along with the Bergen Social Media Addiction Scale,the Internet Gaming Disorder Scale-Short Form,and the Nine-item Adaptability Scale.Results:A four-class model identified by LPA was optimally selected:Class 1(high general mattering,low anti-mattering,low fear of not mattering),Class 2(moderate levels),Class 3(moderate general mattering,high antimattering,high fear of not mattering),and Class 4(low general mattering,low fear of not mattering,moderate anti-mattering).Significant differences were found among these classes in both PIU types(PSMU:F=139.66,p<0.001;PG:F=162.96,p<0.001).The pattern of mean differences consistently showed:Class 3>Class 2>Class 4>Class 1.Class 3 participants demonstrated the highest likelihood of meeting the addiction criteria,Class 2 showed moderate probability,while Classes 1 and 4 exhibited lower probabilities(χ^(2)=113.38 to 408.87,all p<0.001).Additionally,Class 3 reported the lowest adaptability(F=131.67,p<0.001).Conclusion:This study reveals that the unique influence of three ways of assessing feelings of mattering and the fear of not mattering on university students’PIU at the personal level,concluding that these factors are integral to understanding PIU among this demographic.展开更多
Surface-latent heat(LE)and sensible heat(SH)fluxes play a pivotal role in governing hydrological,biological,geochemical,and ecological processes on the land surface in the Tibetan Plateau.However,to accurately assess ...Surface-latent heat(LE)and sensible heat(SH)fluxes play a pivotal role in governing hydrological,biological,geochemical,and ecological processes on the land surface in the Tibetan Plateau.However,to accurately assess and understand the spatial distribution of LE and SH fluxes across different underlying surfaces,it is crucial to verify the validity and reliability of ERA-5,GLDAS,and MODIS data against ground measurements obtained from the Flux Net micrometeorological tower network.This study analyzed the spatial patterns of LE and SH over the Tibetan Plateau using data from ERA-5,GLDAS,and MODIS.The results were compared with ground measurements from Flux Net tower observations on different underlying surfaces,and five statistical parameters(Pearson's r,LR slope,RMSE,MBE,and MAE)were used to validate the data.The results showed that:(1)MODIS LE data and ERA-5 SH data exhibited the closest agreement with ground observations,as indicated by their lowest root mean square error and mean bias area values.(2)The accuracy of ERA-5 SH was the highest in meadows and steppes,while GLDAS SH performed optimally in shrublands.Notably,MODIS LE consistently outperformed the other datasets across all vegetation types.(3)The spatial distribution of LE and SH displayed considerable heterogeneity,contingent upon the specific data sources and underlying surfaces.Notably,there was a contrasting trend between GLDAS and ERA-5,as well as MODIS,in terms of SH distribution in the shrubland.In shrublands and meadows,MODIS SH and LE exhibited more pronounced changes than ERA-5 and GLDAS.Additionally,ERA-5 SH demonstrated the opposite variation in meadow and steppe regions compared to GLDAS and MODIS.展开更多
Physics-informed neural networks(PINNs)are promising to replace conventional mesh-based partial tial differen-equation(PDE)solvers by offering more accurate and flexible PDE solutions.However,PINNs are hampered by the...Physics-informed neural networks(PINNs)are promising to replace conventional mesh-based partial tial differen-equation(PDE)solvers by offering more accurate and flexible PDE solutions.However,PINNs are hampered by the relatively slow convergence and the need to perform additional,potentially expensive training for new PDE parameters.To solve this limitation,we introduce LatentPINN,a framework that utilizes latent representations of the PDE parameters as additional(to the coordinates)inputs into PINNs and allows for training over the distribution of these parameters.Motivated by the recent progress on generative models,we promote using latent diffusion models to learn compressed latent representations of the distribution of PDE parameters as they act as input parameters for NN functional solutions.We use a two-stage training scheme in which,in the first stage,we learn the latent representations for the distribution of PDE parameters.In the second stage,we train a physics-informed neural network over inputs given by randomly drawn samples from the coordinate space within the solution domain and samples from the learned latent representation of the PDE parameters.Considering their importance in capturing evolving interfaces and fronts in various fields,we test the approach on a class of level set equations given,for example,by the nonlinear Eikonal equation.We share results corresponding to three Eikonal parameters(velocity models)sets.The proposed method performs well on new phase velocity models without the need for any additional training.展开更多
Infrared and visible image fusion technology integrates the thermal radiation information of infrared images with the texture details of visible images to generate more informative fused images.However,existing method...Infrared and visible image fusion technology integrates the thermal radiation information of infrared images with the texture details of visible images to generate more informative fused images.However,existing methods often fail to distinguish salient objects from background regions,leading to detail suppression in salient regions due to global fusion strategies.This study presents a mask-guided latent low-rank representation fusion method to address this issue.First,the GrabCut algorithm is employed to extract a saliency mask,distinguishing salient regions from background regions.Then,latent low-rank representation(LatLRR)is applied to extract deep image features,enhancing key information extraction.In the fusion stage,a weighted fusion strategy strengthens infrared thermal information and visible texture details in salient regions,while an average fusion strategy improves background smoothness and stability.Experimental results on the TNO dataset demonstrate that the proposed method achieves superior performance in SPI,MI,Qabf,PSNR,and EN metrics,effectively preserving salient target details while maintaining balanced background information.Compared to state-of-the-art fusion methods,our approach achieves more stable and visually consistent fusion results.The fusion code is available on GitHub at:https://github.com/joyzhen1/Image(accessed on 15 January 2025).展开更多
Efficient and accurate prediction of ocean surface latent heat fluxes is essential for understanding and modeling climate dynamics.Conventional estimation methods have low resolution and lack accuracy.The transformer ...Efficient and accurate prediction of ocean surface latent heat fluxes is essential for understanding and modeling climate dynamics.Conventional estimation methods have low resolution and lack accuracy.The transformer model,with its self-attention mechanism,effectively captures long-range dependencies,leading to a degradation of accuracy over time.Due to the non-linearity and uncertainty of physical processes,the transformer model encounters the problem of error accumulation,leading to a degradation of accuracy over time.To solve this problem,we combine the Data Assimilation(DA)technique with the transformer model and continuously modify the model state to make it closer to the actual observations.In this paper,we propose a deep learning model called TransNetDA,which integrates transformer,convolutional neural network and DA methods.By combining data-driven and DA methods for spatiotemporal prediction,TransNetDA effectively extracts multi-scale spatial features and significantly improves prediction accuracy.The experimental results indicate that the TransNetDA method surpasses traditional techniques in terms of root mean square error and R2 metrics,showcasing its superior performance in predicting latent heat fluxes at the ocean surface.展开更多
This paper analyzes the global competitive landscape of smartphone technological innovation capacity using the latent semantic indexing(LSI)and the vector space model(VSM).It integrates the theory of technological eco...This paper analyzes the global competitive landscape of smartphone technological innovation capacity using the latent semantic indexing(LSI)and the vector space model(VSM).It integrates the theory of technological ecological niches and evaluates four key dimensions:patent quality,energy efficiency engineering,technological modules,and intelligent computing power.The findings reveal that USA has established strong technological barriers through standard-essential patents(SEPs)in wireless communication and integrated circuits.In contrast,Chinese mainland firms rely heavily on fundamental technologies.Qualcomm Inc.in USA and Taiwan Semiconductor Manufacturing Company(TSMC)in Chineses Taiwan have built a comprehensive patent porfolio in energy efficiency engineering.While Chinese mainland faces challenges in advancing dynamic frequency modulation algorithms and high-end manufacturing processes.Huawei Inc.in Chinese mainland leads in 5G module technology but struggles with ecosystem collaboration.Semiconductor manufacturing and radio frequency(RF)components still rely on external suppliers.This highlights the urgent need for innovation in new materials and open'source architectures.To enhance intelligent computing power,Chinese mainland firms must address coordination challenges.They should adopt scenario-driven technological strategies and build a comprehensive ecosystem that includes hardware,operating systems,and developer networks.展开更多
Dear Editor,This letter presents a latent-factorization-of-tensors(LFT)-incorporated battery cycle life prediction framework.Data-driven prognosis and health management(PHM)for battery pack(BP)can boost the safety and...Dear Editor,This letter presents a latent-factorization-of-tensors(LFT)-incorporated battery cycle life prediction framework.Data-driven prognosis and health management(PHM)for battery pack(BP)can boost the safety and sustainability of a battery management system(BMS),which relies heavily on the quality of the measured BP data like the voltage(V),current(I),and temperature(T).展开更多
Objectives:SocialNetworkAddiction(SNA)is becoming increasingly prevalent among college students;however,there remains a lack of consensus regarding the measurement tools and their optimal cutoff score.This study aims ...Objectives:SocialNetworkAddiction(SNA)is becoming increasingly prevalent among college students;however,there remains a lack of consensus regarding the measurement tools and their optimal cutoff score.This study aims to validate the 21-item SocialNetwork Addiction Scale-Chinese(SNAS-C)in its Chinese version and to determine its optimal cutoff score for identifying potential SNA cases within the college student population.Methods:A crosssectional survey was conducted,recruiting 3387 college students.Latent profile analysis(LPA)and receiver operating characteristic(ROC)curve analysis were employed to establish the optimal cutoff score for the validated 21-item SNAS-C.Results:Three profile models were selected based on multiple statistical criteria,classifying participants into low-risk,moderate-risk,and high-risk groups.The highest-risk group was defined as“positive”for SNA,while the remaining groups were considered“negative”,serving as the reference standard for ROC analysis.The optimal cutoff score was determined to be 72(sensitivity:98.2%,specificity:96.86%),with an overall classification accuracy of 97.0%.The“positive”group reported significantly higher frequency of social network usage,greater digitalmedia dependence scores,and a higher incidence of network addiction.Conclusion:This study identified the optimal cutoff score for the SNAS-C as≥72,demonstrating high sensitivity,specificity,and diagnostic accuracy.This threshold effectively distinguishes between high-risk and low-risk SNA.展开更多
Water is the most abundant liquid on the surface of the earth. It is a liquid whose properties are quite surprising, both as a pure liquid and as a solvent. Water is a very cohesive liquid: its melting and vaporizatio...Water is the most abundant liquid on the surface of the earth. It is a liquid whose properties are quite surprising, both as a pure liquid and as a solvent. Water is a very cohesive liquid: its melting and vaporization temperatures are very high for a liquid that is neither ionic nor metallic, and whose molar mass is low. Thus, water remains liquid at atmospheric pressure up to 100C while similar molecules such as H2S, H2Se, H2Te for example would give a vaporization temperature close to 80C. This cohesion is in fact ensured by hydrogen bonds between water molecules. This type of bonds between neighboring molecules, hydrogen bonds, is quite often found in chemistry [1] [2]. Any change in the state of aggregation of a substance occurs with the absorption or release of a certain amount of latent heat of transformation. Latent heat of fusion, vaporization or sublimation is the ratio of the energy supplied as heat to the mass of the substance that is melted, vaporized or sublimated. As a result of the reversibility of the processes, the fusion heat is equal to the heat released in the reverse process: crystallization and solidification heat. And likewise the heat of vaporization is equal to the heat of condensation. This equality of heat is often used to determine experimentally either of these quantities. There are two main measurement methods: 1) Direct measurement using the calorimeter, 2) Indirect measure based on the use of the VantHoff relationship. The objective of this work is to measure the latent heat of water vaporization and verify the compatibility of the experimental values with the values given by the tables using the indirect method.展开更多
BACKGROUND Non-suicidal self-injury(NSSI)is common among adolescents and frequently cooccurs with depression.Understanding the distinct patterns of NSSI behaviors,along with their associated risk and protective factor...BACKGROUND Non-suicidal self-injury(NSSI)is common among adolescents and frequently cooccurs with depression.Understanding the distinct patterns of NSSI behaviors,along with their associated risk and protective factors,is crucial for developing effective interventions.AIM To classify NSSI behaviors and examine interactions between risk and resilience factors in Chinese adolescents.METHODS A cross-sectional study involving 3967 Chinese students(51.7%female,mean age 13.58±2.24 years)who completed questionnaires on parenting styles,bullying,childhood maltreatment,depression,resilience,and NSSI.Latent profile analysis(LPA)was used to identify NSSI subtypes,and network analysis explored interactions between risk and resilience factors.RESULTS Three NSSI subtypes were identified:NSSI with depression(18.8%),NSSI without depression(12.3%),and neither(68.9%).Bullying was the central risk factor across subtypes,while emotional control and family support were key protective factors.Statistical analyses showed significant differences between groups(P<0.001).CONCLUSION This study identified three NSSI subtypes among Chinese adolescents.Bullying emerged as a central risk factor,while emotional control and family support were key protective factors.Targeting these areas may help reduce NSSI behaviors in this population.展开更多
Psychiatric comorbidity is common in symptombased diagnoses like autism spectrum disorder(ASD),attention/deficit hyper-activity disorder(ADHD),and obsessivecompulsive disorder(OCD).However,these co-occurring symptoms ...Psychiatric comorbidity is common in symptombased diagnoses like autism spectrum disorder(ASD),attention/deficit hyper-activity disorder(ADHD),and obsessivecompulsive disorder(OCD).However,these co-occurring symptoms mediated by shared and/or distinct neural mechanisms are difficult to profile at the individual level.Capitalizing on unsupervised machine learning with a hierarchical Bayesian framework,we derived latent disease factors from resting-state functional connectivity data in a hybrid cohort of ASD and ADHD and delineated individual associations with dimensional symptoms based on canonical correlation analysis.Models based on the same factors generalized to previously unseen individuals in a subclinical cohort and one local OCD database with a subset of patients undergoing neurosurgical intervention.Four factors,identified as variably co-expressed in each patient,were significantly correlated with distinct symptom domains(r=–0.26–0.53,P<0.05):behavioral regulation(Factor-1),communication(Factor-2),anxiety(Factor-3),adaptive behaviors(Factor-4).Moreover,we demonstrated Factor-1 expressed in patients with OCD and Factor-3 expressed in participants with anxiety,at the degree to which factor expression was significantly predictive of individual symptom scores(r=0.18–0.5,P<0.01).Importantly,peri-intervention changes in Factor-1 of OCD were associated with variable treatment outcomes(r=0.39,P<0.05).Our results indicate that these data-derived latent disease factors quantify individual factor expression to inform dimensional symptom and treatment outcomes across cohorts,which may promote quantitative psychiatric diagnosis and personalized intervention.展开更多
Multimodal medical image fusion can help physicians provide more accurate treatment plans for patients, as unimodal images provide limited valid information. To address the insufficient ability of traditional medical ...Multimodal medical image fusion can help physicians provide more accurate treatment plans for patients, as unimodal images provide limited valid information. To address the insufficient ability of traditional medical image fusion solutions to protect image details and significant information, a new multimodality medical image fusion method(NSST-PAPCNNLatLRR) is proposed in this paper. Firstly, the high and low-frequency sub-band coefficients are obtained by decomposing the source image using NSST. Then, the latent low-rank representation algorithm is used to process the low-frequency sub-band coefficients;An improved PAPCNN algorithm is also proposed for the fusion of high-frequency sub-band coefficients. The improved PAPCNN model was based on the automatic setting of the parameters, and the optimal method was configured for the time decay factor αe. The experimental results show that, in comparison with the five mainstream fusion algorithms, the new algorithm has significantly improved the visual effect over the comparison algorithm,enhanced the ability to characterize important information in images, and further improved the ability to protect the detailed information;the new algorithm has achieved at least four firsts in six objective indexes.展开更多
Latent tuberculosis infection(LTBI)has become a major source of active tuberculosis(ATB).Although the tuberculin skin test and interferon-gamma release assay can be used to diagnose LTBI,these methods can only differe...Latent tuberculosis infection(LTBI)has become a major source of active tuberculosis(ATB).Although the tuberculin skin test and interferon-gamma release assay can be used to diagnose LTBI,these methods can only differentiate infected individuals from healthy ones but cannot discriminate between LTBI and ATB.Thus,the diagnosis of LTBI faces many challenges,such as the lack of effective biomarkers from Mycobacterium tuberculosis(MTB)for distinguishing LTBI,the low diagnostic efficacy of biomarkers derived from the human host,and the absence of a gold standard to differentiate between LTBI and ATB.Sputum culture,as the gold standard for diagnosing tuberculosis,is time-consuming and cannot distinguish between ATB and LTBI.In this article,we review the pathogenesis of MTB and the immune mechanisms of the host in LTBI,including the innate and adaptive immune responses,multiple immune evasion mechanisms of MTB,and epigenetic regulation.Based on this knowledge,we summarize the current status and challenges in diagnosing LTBI and present the application of machine learning(ML)in LTBI diagnosis,as well as the advantages and limitations of ML in this context.Finally,we discuss the future development directions of ML applied to LTBI diagnosis.展开更多
Bayesian networks are a powerful class of graphical decision models used to represent causal relationships among variables.However,the reliability and integrity of learned Bayesian network models are highly dependent ...Bayesian networks are a powerful class of graphical decision models used to represent causal relationships among variables.However,the reliability and integrity of learned Bayesian network models are highly dependent on the quality of incoming data streams.One of the primary challenges with Bayesian networks is their vulnerability to adversarial data poisoning attacks,wherein malicious data is injected into the training dataset to negatively influence the Bayesian network models and impair their performance.In this research paper,we propose an efficient framework for detecting data poisoning attacks against Bayesian network structure learning algorithms.Our framework utilizes latent variables to quantify the amount of belief between every two nodes in each causal model over time.We use our innovative methodology to tackle an important issue with data poisoning assaults in the context of Bayesian networks.With regard to four different forms of data poisoning attacks,we specifically aim to strengthen the security and dependability of Bayesian network structure learning techniques,such as the PC algorithm.By doing this,we explore the complexity of this area and offer workablemethods for identifying and reducing these sneaky dangers.Additionally,our research investigates one particular use case,the“Visit to Asia Network.”The practical consequences of using uncertainty as a way to spot cases of data poisoning are explored in this inquiry,which is of utmost relevance.Our results demonstrate the promising efficacy of latent variables in detecting and mitigating the threat of data poisoning attacks.Additionally,our proposed latent-based framework proves to be sensitive in detecting malicious data poisoning attacks in the context of stream data.展开更多
Efficient exploration in complex coordination tasks has been considered a challenging problem in multi-agent reinforcement learning(MARL). It is significantly more difficult for those tasks with latent variables that ...Efficient exploration in complex coordination tasks has been considered a challenging problem in multi-agent reinforcement learning(MARL). It is significantly more difficult for those tasks with latent variables that agents cannot directly observe. However, most of the existing latent variable discovery methods lack a clear representation of latent variables and an effective evaluation of the influence of latent variables on the agent. In this paper, we propose a new MARL algorithm based on the soft actor-critic method for complex continuous control tasks with confounders. It is called the multi-agent soft actor-critic with latent variable(MASAC-LV) algorithm, which uses variational inference theory to infer the compact latent variables representation space from a large amount of offline experience.Besides, we derive the counterfactual policy whose input has no latent variables and quantify the difference between the actual policy and the counterfactual policy via a distance function. This quantified difference is considered an intrinsic motivation that gives additional rewards based on how much the latent variable affects each agent. The proposed algorithm is evaluated on two collaboration tasks with confounders, and the experimental results demonstrate the effectiveness of MASAC-LV compared to other baseline algorithms.展开更多
基金supported by Higher Institution Centre of Excellence(HICoE)Grant(A305-KR-AKH002-0000000278-K134)from the Ministry of Higher Education,Malaysia.
文摘Objective:To assess aptamer-based assays for diagnosing latent tuberculosis infection(LTBI).Methods:Literature from Medline,ScienceDirect,and Scopus,covering publications from January 1,2012,to December 31,2023,was examined.This review evaluates different aptamers,biomarkers,sample types,sample sizes,reference assays,and the assays'sensitivity and specificity.By using the Quality Assessment of Diagnostic Accuracy Studies 2,the risk of bias in each study was evaluated.Results:Aptamer-based assays generally showed a sensitivity of 90%(95%CI:75%-100%)and specificity of 90%(95%CI:50%-100%),where optical aptasensor showed the highest sensitivity and specificity at 100%.Serum samples were frequently used to enhance antigen detectability,improving the assay’s performance.Meanwhile,HspX was the most studied biomarker,followed by MPT64,and IFN-γ.Conclusions:Aptamer-based assays could be reliable alternatives to current LTBI detection methods,but further research is needed to validate their clinical efficacy.
基金supported by the“333 High-Level Talents Training Project”of Jiangsu province(No.BRA2020069)。
文摘Objectives:The study aimed to explore social alienation types in patients with inflammatorybowel diseases(IBD)and identify influencingfactors.Methods:This cross-sectional study was conducted using purposive sampling among patients with IBD from July 2022 to July 2023.Patients were assessed using the Generalized Social Alienation Scale(GSAS),the Brief Illness Perception Questionnaire(B-IPQ),the Hospital Anxiety and Depression Scale(HADS),and the Medical Coping Modes Questionnaire(MCMQ).Demographic and disease-related characteristics were also collected.Latent profileanalysis(LPA)was used to identify potential subgroups of social alienation.Univariate analysis and multicollinearity analysis were conducted to explore the influencing factors,followed by multiple regression analysis to evaluate the effect of influencingfactors on social alienation.Results:Three distinct profilesof social alienation were identified:integrated-low alienation group(n=61,20.20%),accommodative-moderate alienation group(n=195,64.57%),and maladaptive-high alienation group(n=46,15.23%).Seven characteristics were associated with the profile’smembership:self-perceived financialstress,malnutrition risk,disease duration,illness comprehensibility,anxiety,depression,and acceptance-resignation coping mode.Conclusions:Patients with IBD were categorized into three subgroups based on social alienation levels.Financial stress,malnutrition risk,disease duration,illness comprehensibility,anxiety,depression,and acceptance-resignation coping mode were key predictors of the subgroup membership.Targeted interventions should be developed to mitigate the negative effects of social alienation,with a focus on improving illness perception,alleviating anxiety and depression,and promoting effective coping strategies.
基金supported by the China Postdoctoral Science Foundation(No.2014m561331)Science and Technology Research Project of Heilongjiang Provincial Education Department(No.12521073)National Natural Science Youth Fund(No.61300115).
文摘In the field of organic synthesis,the core objective of retrosynthetic methods is to deduce possible synthetic routes and precursor molecules for complex target molecules.Traditional retrosynthetic methods,such as template-based retrosynthesis,have high accuracy and interpretability in specific types of reactions but are limited by the scope of the template library,making it difficult to adapt to new or uncommon reaction types.Moreover,sequence-to-sequence retrosynthetic prediction methods,although they enhance the flexibility of prediction,often overlook the complexity of molecular graph structures and the actual interactions between atoms,which limits the accuracy and reliability of the predictions.To address these limitations,this paper proposes a Molecular Retrosynthesis Top-k Prediction based on the Latent Generation Process(MRLGP)that uses latent variables from graph neural networks to model the generation process and produce diverse set of reactants.Utilising an encoding method based on Graphormer,the authors have also introduced topology-aware positional encoding to better capture the interactions between atomic nodes in the molecular graph structure,thereby more accurately simulating the retrosynthetic process.The MRLGP model significantly enhances the accuracy and diversity of predictions by correlating discrete latent variables with the reactant generation process and progressively constructing molecular graphs using a variational autoregressive decoder.Experimental results on benchmark datasets such as USPTO-50k,USPTO-Full,and USPTO-DIVERSE demonstrate that MRLGP outperforms baseline models on multiple Top-k evaluation metrics.Additionally,ablation experiments conducted on the USPTO-50K dataset further validate the effectiveness of the methods used in the encoder and decoder parts of the model.
基金the General project of the“14th Five-Year Plan”of Education Science of Gansu Province in 2023:“Research on the Integration Strategy of Rural Primary School Labor Education and Other Disciplines under the Background of Rural Revitalization”(Project number:GS[2023]GHB1420)Gansu Provincial University Curriculum Ideological and Political Demonstration Project“Exploration and Practice of Ideological and Political Courses in Pre-School EducationMajors from the Perspective of Three Educations:A Case study of Pre-School EducationHistory”(Project No.:GSkcsz-2021-094).
文摘Objectives:Adolescents’aggression is widely studied,the underlying heterogeneity of aggression among rural Chinese boarding students remains unexplored.This study investigates the latent profiles of Chinese rural boarding junior high school students’aggression and its correlations with moral disengagement and preference for solitude.Methods:A cross-sectional survey was conducted from 04–22 April 2022,using a convenient sampling method among 516 junior high school students from six Chinese rural boarding schools.The survey included the Aggression Questionnaire,the Moral Disengagement Scale(MDS),and the Preference for Solitude Scale(PSS).Results:Participants were divided into three latent profiles:low(36.0%),medium(50.9%),and high aggression levels(13.1%).Compared with low aggression,students who felt left-behind(minors who stay in the rural areas while one or both parents migrated to the urban areas for the work)accounted for a larger proportion in the medium aggression profile.The higher the grade level and the lower the educational level of the students’parents,the greater proportion of students in the medium and high aggression profiles.Additionally,students with high moral disengagement and preference for solitude showed a significant association with the medium aggression and high aggression profiles.Conclusions:The results demonstrate the significant group heterogeneity of aggression groups in Chinese rural boarding junior high school students.Targeted prevention and intervention measures can be carried out according to feeling left-behind,grade level,parents’education,and MDS and PSS scores.
基金supported by a special grant from the Taishan Scholars Project(Project No.tsqn202211130).
文摘Background:This study addresses the pressing need to understand the nuanced relationship between‘mattering’—the perception of being significant to others—and problematic internet use(PIU)among university students.Unlike previous research that has primarily employed variable-centered approaches,this study first adopts a person-centered approach using Latent Profile Analysis(LPA)to identify distinct mattering profiles.Subsequently,through variable-centered analyses,these profiles are examined in relation to different types of PIU—specifically problematic social media use(PSMU)and problematic gaming(PG)—as well as adaptability.Methods:Data were collected from 3587 university students across 19 universities in China.Participants completed three mattering-related scales(General Mattering Scale,Anti-Mattering Scale,and Fear of Not Mattering Inventory),along with the Bergen Social Media Addiction Scale,the Internet Gaming Disorder Scale-Short Form,and the Nine-item Adaptability Scale.Results:A four-class model identified by LPA was optimally selected:Class 1(high general mattering,low anti-mattering,low fear of not mattering),Class 2(moderate levels),Class 3(moderate general mattering,high antimattering,high fear of not mattering),and Class 4(low general mattering,low fear of not mattering,moderate anti-mattering).Significant differences were found among these classes in both PIU types(PSMU:F=139.66,p<0.001;PG:F=162.96,p<0.001).The pattern of mean differences consistently showed:Class 3>Class 2>Class 4>Class 1.Class 3 participants demonstrated the highest likelihood of meeting the addiction criteria,Class 2 showed moderate probability,while Classes 1 and 4 exhibited lower probabilities(χ^(2)=113.38 to 408.87,all p<0.001).Additionally,Class 3 reported the lowest adaptability(F=131.67,p<0.001).Conclusion:This study reveals that the unique influence of three ways of assessing feelings of mattering and the fear of not mattering on university students’PIU at the personal level,concluding that these factors are integral to understanding PIU among this demographic.
基金funded by the West Light Scholar of the Chinese Academy of Sciences(xbzg-zdsys-202202)the Natural Science Foundation of Henan(Grant No.232300420165)Integrated Scientific Investigation of the North-South Transitional Zone of China(2017FY100900)。
文摘Surface-latent heat(LE)and sensible heat(SH)fluxes play a pivotal role in governing hydrological,biological,geochemical,and ecological processes on the land surface in the Tibetan Plateau.However,to accurately assess and understand the spatial distribution of LE and SH fluxes across different underlying surfaces,it is crucial to verify the validity and reliability of ERA-5,GLDAS,and MODIS data against ground measurements obtained from the Flux Net micrometeorological tower network.This study analyzed the spatial patterns of LE and SH over the Tibetan Plateau using data from ERA-5,GLDAS,and MODIS.The results were compared with ground measurements from Flux Net tower observations on different underlying surfaces,and five statistical parameters(Pearson's r,LR slope,RMSE,MBE,and MAE)were used to validate the data.The results showed that:(1)MODIS LE data and ERA-5 SH data exhibited the closest agreement with ground observations,as indicated by their lowest root mean square error and mean bias area values.(2)The accuracy of ERA-5 SH was the highest in meadows and steppes,while GLDAS SH performed optimally in shrublands.Notably,MODIS LE consistently outperformed the other datasets across all vegetation types.(3)The spatial distribution of LE and SH displayed considerable heterogeneity,contingent upon the specific data sources and underlying surfaces.Notably,there was a contrasting trend between GLDAS and ERA-5,as well as MODIS,in terms of SH distribution in the shrubland.In shrublands and meadows,MODIS SH and LE exhibited more pronounced changes than ERA-5 and GLDAS.Additionally,ERA-5 SH demonstrated the opposite variation in meadow and steppe regions compared to GLDAS and MODIS.
基金King Abdullah University of Science and Technol-ogy(KAUST)for supporting this research and the Seismic Wave Anal-ysis group for the supportive and encouraging environment.
文摘Physics-informed neural networks(PINNs)are promising to replace conventional mesh-based partial tial differen-equation(PDE)solvers by offering more accurate and flexible PDE solutions.However,PINNs are hampered by the relatively slow convergence and the need to perform additional,potentially expensive training for new PDE parameters.To solve this limitation,we introduce LatentPINN,a framework that utilizes latent representations of the PDE parameters as additional(to the coordinates)inputs into PINNs and allows for training over the distribution of these parameters.Motivated by the recent progress on generative models,we promote using latent diffusion models to learn compressed latent representations of the distribution of PDE parameters as they act as input parameters for NN functional solutions.We use a two-stage training scheme in which,in the first stage,we learn the latent representations for the distribution of PDE parameters.In the second stage,we train a physics-informed neural network over inputs given by randomly drawn samples from the coordinate space within the solution domain and samples from the learned latent representation of the PDE parameters.Considering their importance in capturing evolving interfaces and fronts in various fields,we test the approach on a class of level set equations given,for example,by the nonlinear Eikonal equation.We share results corresponding to three Eikonal parameters(velocity models)sets.The proposed method performs well on new phase velocity models without the need for any additional training.
基金supported by Universiti Teknologi MARA through UiTM MyRA Research Grant,600-RMC 5/3/GPM(053/2022).
文摘Infrared and visible image fusion technology integrates the thermal radiation information of infrared images with the texture details of visible images to generate more informative fused images.However,existing methods often fail to distinguish salient objects from background regions,leading to detail suppression in salient regions due to global fusion strategies.This study presents a mask-guided latent low-rank representation fusion method to address this issue.First,the GrabCut algorithm is employed to extract a saliency mask,distinguishing salient regions from background regions.Then,latent low-rank representation(LatLRR)is applied to extract deep image features,enhancing key information extraction.In the fusion stage,a weighted fusion strategy strengthens infrared thermal information and visible texture details in salient regions,while an average fusion strategy improves background smoothness and stability.Experimental results on the TNO dataset demonstrate that the proposed method achieves superior performance in SPI,MI,Qabf,PSNR,and EN metrics,effectively preserving salient target details while maintaining balanced background information.Compared to state-of-the-art fusion methods,our approach achieves more stable and visually consistent fusion results.The fusion code is available on GitHub at:https://github.com/joyzhen1/Image(accessed on 15 January 2025).
基金The National Natural Science Foundation of China under contract Nos 42176011 and 61931025the Fundamental Research Funds for the Central Universities of China under contract No.24CX03001A.
文摘Efficient and accurate prediction of ocean surface latent heat fluxes is essential for understanding and modeling climate dynamics.Conventional estimation methods have low resolution and lack accuracy.The transformer model,with its self-attention mechanism,effectively captures long-range dependencies,leading to a degradation of accuracy over time.Due to the non-linearity and uncertainty of physical processes,the transformer model encounters the problem of error accumulation,leading to a degradation of accuracy over time.To solve this problem,we combine the Data Assimilation(DA)technique with the transformer model and continuously modify the model state to make it closer to the actual observations.In this paper,we propose a deep learning model called TransNetDA,which integrates transformer,convolutional neural network and DA methods.By combining data-driven and DA methods for spatiotemporal prediction,TransNetDA effectively extracts multi-scale spatial features and significantly improves prediction accuracy.The experimental results indicate that the TransNetDA method surpasses traditional techniques in terms of root mean square error and R2 metrics,showcasing its superior performance in predicting latent heat fluxes at the ocean surface.
基金supported in part by the National Social Science Foundation of China(No.20BGL203).
文摘This paper analyzes the global competitive landscape of smartphone technological innovation capacity using the latent semantic indexing(LSI)and the vector space model(VSM).It integrates the theory of technological ecological niches and evaluates four key dimensions:patent quality,energy efficiency engineering,technological modules,and intelligent computing power.The findings reveal that USA has established strong technological barriers through standard-essential patents(SEPs)in wireless communication and integrated circuits.In contrast,Chinese mainland firms rely heavily on fundamental technologies.Qualcomm Inc.in USA and Taiwan Semiconductor Manufacturing Company(TSMC)in Chineses Taiwan have built a comprehensive patent porfolio in energy efficiency engineering.While Chinese mainland faces challenges in advancing dynamic frequency modulation algorithms and high-end manufacturing processes.Huawei Inc.in Chinese mainland leads in 5G module technology but struggles with ecosystem collaboration.Semiconductor manufacturing and radio frequency(RF)components still rely on external suppliers.This highlights the urgent need for innovation in new materials and open'source architectures.To enhance intelligent computing power,Chinese mainland firms must address coordination challenges.They should adopt scenario-driven technological strategies and build a comprehensive ecosystem that includes hardware,operating systems,and developer networks.
文摘Dear Editor,This letter presents a latent-factorization-of-tensors(LFT)-incorporated battery cycle life prediction framework.Data-driven prognosis and health management(PHM)for battery pack(BP)can boost the safety and sustainability of a battery management system(BMS),which relies heavily on the quality of the measured BP data like the voltage(V),current(I),and temperature(T).
基金supported by the National Natural Science Foundation of China(Grant No.72364006).
文摘Objectives:SocialNetworkAddiction(SNA)is becoming increasingly prevalent among college students;however,there remains a lack of consensus regarding the measurement tools and their optimal cutoff score.This study aims to validate the 21-item SocialNetwork Addiction Scale-Chinese(SNAS-C)in its Chinese version and to determine its optimal cutoff score for identifying potential SNA cases within the college student population.Methods:A crosssectional survey was conducted,recruiting 3387 college students.Latent profile analysis(LPA)and receiver operating characteristic(ROC)curve analysis were employed to establish the optimal cutoff score for the validated 21-item SNAS-C.Results:Three profile models were selected based on multiple statistical criteria,classifying participants into low-risk,moderate-risk,and high-risk groups.The highest-risk group was defined as“positive”for SNA,while the remaining groups were considered“negative”,serving as the reference standard for ROC analysis.The optimal cutoff score was determined to be 72(sensitivity:98.2%,specificity:96.86%),with an overall classification accuracy of 97.0%.The“positive”group reported significantly higher frequency of social network usage,greater digitalmedia dependence scores,and a higher incidence of network addiction.Conclusion:This study identified the optimal cutoff score for the SNAS-C as≥72,demonstrating high sensitivity,specificity,and diagnostic accuracy.This threshold effectively distinguishes between high-risk and low-risk SNA.
文摘Water is the most abundant liquid on the surface of the earth. It is a liquid whose properties are quite surprising, both as a pure liquid and as a solvent. Water is a very cohesive liquid: its melting and vaporization temperatures are very high for a liquid that is neither ionic nor metallic, and whose molar mass is low. Thus, water remains liquid at atmospheric pressure up to 100C while similar molecules such as H2S, H2Se, H2Te for example would give a vaporization temperature close to 80C. This cohesion is in fact ensured by hydrogen bonds between water molecules. This type of bonds between neighboring molecules, hydrogen bonds, is quite often found in chemistry [1] [2]. Any change in the state of aggregation of a substance occurs with the absorption or release of a certain amount of latent heat of transformation. Latent heat of fusion, vaporization or sublimation is the ratio of the energy supplied as heat to the mass of the substance that is melted, vaporized or sublimated. As a result of the reversibility of the processes, the fusion heat is equal to the heat released in the reverse process: crystallization and solidification heat. And likewise the heat of vaporization is equal to the heat of condensation. This equality of heat is often used to determine experimentally either of these quantities. There are two main measurement methods: 1) Direct measurement using the calorimeter, 2) Indirect measure based on the use of the VantHoff relationship. The objective of this work is to measure the latent heat of water vaporization and verify the compatibility of the experimental values with the values given by the tables using the indirect method.
基金Supported by Yunnan Province High-Level Health Technical Talents,Leading Talents,No.L-2019011.
文摘BACKGROUND Non-suicidal self-injury(NSSI)is common among adolescents and frequently cooccurs with depression.Understanding the distinct patterns of NSSI behaviors,along with their associated risk and protective factors,is crucial for developing effective interventions.AIM To classify NSSI behaviors and examine interactions between risk and resilience factors in Chinese adolescents.METHODS A cross-sectional study involving 3967 Chinese students(51.7%female,mean age 13.58±2.24 years)who completed questionnaires on parenting styles,bullying,childhood maltreatment,depression,resilience,and NSSI.Latent profile analysis(LPA)was used to identify NSSI subtypes,and network analysis explored interactions between risk and resilience factors.RESULTS Three NSSI subtypes were identified:NSSI with depression(18.8%),NSSI without depression(12.3%),and neither(68.9%).Bullying was the central risk factor across subtypes,while emotional control and family support were key protective factors.Statistical analyses showed significant differences between groups(P<0.001).CONCLUSION This study identified three NSSI subtypes among Chinese adolescents.Bullying emerged as a central risk factor,while emotional control and family support were key protective factors.Targeting these areas may help reduce NSSI behaviors in this population.
基金supported by the National Natural Science Foundation(82151303)the National Key R&D Program of China(2021ZD0204002)+3 种基金the Key-Area Research and Development Program of Guangdong Province(2019B030335001)Shanghai Municipal Science and Technology Major Project(2018SHZDZX05)Strategic Priority Research Program of Chinese Academy of Science(XDB32000000)supported in part by the Postdoctoral Fellowship of the Peking-Tsinghua Center for Life Sciences.GSP data were provided by the Brain Genomics Superstruct Project of Harvard University and the Massachusetts General Hospital,(Principal Investigators:Randy Buckner,Joshua Roffman,and Jordan Smoller),with support from the Center for Brain Science Neuroinformatics Research Group,the Athinoula A.Martinos Center for Biomedical Imaging,and GSP Open Access Documentation the Center for Human Genetic Research.20 individual investigators at Harvard and MGH generously contributed data to the overall project.
文摘Psychiatric comorbidity is common in symptombased diagnoses like autism spectrum disorder(ASD),attention/deficit hyper-activity disorder(ADHD),and obsessivecompulsive disorder(OCD).However,these co-occurring symptoms mediated by shared and/or distinct neural mechanisms are difficult to profile at the individual level.Capitalizing on unsupervised machine learning with a hierarchical Bayesian framework,we derived latent disease factors from resting-state functional connectivity data in a hybrid cohort of ASD and ADHD and delineated individual associations with dimensional symptoms based on canonical correlation analysis.Models based on the same factors generalized to previously unseen individuals in a subclinical cohort and one local OCD database with a subset of patients undergoing neurosurgical intervention.Four factors,identified as variably co-expressed in each patient,were significantly correlated with distinct symptom domains(r=–0.26–0.53,P<0.05):behavioral regulation(Factor-1),communication(Factor-2),anxiety(Factor-3),adaptive behaviors(Factor-4).Moreover,we demonstrated Factor-1 expressed in patients with OCD and Factor-3 expressed in participants with anxiety,at the degree to which factor expression was significantly predictive of individual symptom scores(r=0.18–0.5,P<0.01).Importantly,peri-intervention changes in Factor-1 of OCD were associated with variable treatment outcomes(r=0.39,P<0.05).Our results indicate that these data-derived latent disease factors quantify individual factor expression to inform dimensional symptom and treatment outcomes across cohorts,which may promote quantitative psychiatric diagnosis and personalized intervention.
基金funded by the National Natural Science Foundation of China,grant number 61302188.
文摘Multimodal medical image fusion can help physicians provide more accurate treatment plans for patients, as unimodal images provide limited valid information. To address the insufficient ability of traditional medical image fusion solutions to protect image details and significant information, a new multimodality medical image fusion method(NSST-PAPCNNLatLRR) is proposed in this paper. Firstly, the high and low-frequency sub-band coefficients are obtained by decomposing the source image using NSST. Then, the latent low-rank representation algorithm is used to process the low-frequency sub-band coefficients;An improved PAPCNN algorithm is also proposed for the fusion of high-frequency sub-band coefficients. The improved PAPCNN model was based on the automatic setting of the parameters, and the optimal method was configured for the time decay factor αe. The experimental results show that, in comparison with the five mainstream fusion algorithms, the new algorithm has significantly improved the visual effect over the comparison algorithm,enhanced the ability to characterize important information in images, and further improved the ability to protect the detailed information;the new algorithm has achieved at least four firsts in six objective indexes.
文摘Latent tuberculosis infection(LTBI)has become a major source of active tuberculosis(ATB).Although the tuberculin skin test and interferon-gamma release assay can be used to diagnose LTBI,these methods can only differentiate infected individuals from healthy ones but cannot discriminate between LTBI and ATB.Thus,the diagnosis of LTBI faces many challenges,such as the lack of effective biomarkers from Mycobacterium tuberculosis(MTB)for distinguishing LTBI,the low diagnostic efficacy of biomarkers derived from the human host,and the absence of a gold standard to differentiate between LTBI and ATB.Sputum culture,as the gold standard for diagnosing tuberculosis,is time-consuming and cannot distinguish between ATB and LTBI.In this article,we review the pathogenesis of MTB and the immune mechanisms of the host in LTBI,including the innate and adaptive immune responses,multiple immune evasion mechanisms of MTB,and epigenetic regulation.Based on this knowledge,we summarize the current status and challenges in diagnosing LTBI and present the application of machine learning(ML)in LTBI diagnosis,as well as the advantages and limitations of ML in this context.Finally,we discuss the future development directions of ML applied to LTBI diagnosis.
文摘Bayesian networks are a powerful class of graphical decision models used to represent causal relationships among variables.However,the reliability and integrity of learned Bayesian network models are highly dependent on the quality of incoming data streams.One of the primary challenges with Bayesian networks is their vulnerability to adversarial data poisoning attacks,wherein malicious data is injected into the training dataset to negatively influence the Bayesian network models and impair their performance.In this research paper,we propose an efficient framework for detecting data poisoning attacks against Bayesian network structure learning algorithms.Our framework utilizes latent variables to quantify the amount of belief between every two nodes in each causal model over time.We use our innovative methodology to tackle an important issue with data poisoning assaults in the context of Bayesian networks.With regard to four different forms of data poisoning attacks,we specifically aim to strengthen the security and dependability of Bayesian network structure learning techniques,such as the PC algorithm.By doing this,we explore the complexity of this area and offer workablemethods for identifying and reducing these sneaky dangers.Additionally,our research investigates one particular use case,the“Visit to Asia Network.”The practical consequences of using uncertainty as a way to spot cases of data poisoning are explored in this inquiry,which is of utmost relevance.Our results demonstrate the promising efficacy of latent variables in detecting and mitigating the threat of data poisoning attacks.Additionally,our proposed latent-based framework proves to be sensitive in detecting malicious data poisoning attacks in the context of stream data.
基金supported in part by the National Natural Science Foundation of China (62136008,62236002,61921004,62173251,62103104)the “Zhishan” Scholars Programs of Southeast Universitythe Fundamental Research Funds for the Central Universities (2242023K30034)。
文摘Efficient exploration in complex coordination tasks has been considered a challenging problem in multi-agent reinforcement learning(MARL). It is significantly more difficult for those tasks with latent variables that agents cannot directly observe. However, most of the existing latent variable discovery methods lack a clear representation of latent variables and an effective evaluation of the influence of latent variables on the agent. In this paper, we propose a new MARL algorithm based on the soft actor-critic method for complex continuous control tasks with confounders. It is called the multi-agent soft actor-critic with latent variable(MASAC-LV) algorithm, which uses variational inference theory to infer the compact latent variables representation space from a large amount of offline experience.Besides, we derive the counterfactual policy whose input has no latent variables and quantify the difference between the actual policy and the counterfactual policy via a distance function. This quantified difference is considered an intrinsic motivation that gives additional rewards based on how much the latent variable affects each agent. The proposed algorithm is evaluated on two collaboration tasks with confounders, and the experimental results demonstrate the effectiveness of MASAC-LV compared to other baseline algorithms.
