Time series clustering is a challenging problem due to the large-volume,high-dimensional,and warping characteristics of time series data.Traditional clustering methods often use a single criterion or distance measure,...Time series clustering is a challenging problem due to the large-volume,high-dimensional,and warping characteristics of time series data.Traditional clustering methods often use a single criterion or distance measure,which may not capture all the features of the data.This paper proposes a novel method for time series clustering based on evolutionary multi-tasking optimization,termed i-MFEA,which uses an improved multifactorial evolutionary algorithm to optimize multiple clustering tasks simultaneously,each with a different validity index or distance measure.Therefore,i-MFEA can produce diverse and robust clustering solutions that satisfy various preferences of decision-makers.Experiments on two artificial datasets show that i-MFEA outperforms single-objective evolutionary algorithms and traditional clustering methods in terms of convergence speed and clustering quality.The paper also discusses how i-MFEA can address two long-standing issues in time series clustering:the choice of appropriate similarity measure and the number of clusters.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
Cooperative multi-agent reinforcement learning(MARL)is a key technology for enabling cooperation in complex multi-agent systems.It has achieved remarkable progress in areas such as gaming,autonomous driving,and multi-...Cooperative multi-agent reinforcement learning(MARL)is a key technology for enabling cooperation in complex multi-agent systems.It has achieved remarkable progress in areas such as gaming,autonomous driving,and multi-robot control.Empowering cooperative MARL with multi-task decision-making capabilities is expected to further broaden its application scope.In multi-task scenarios,cooperative MARL algorithms need to address 3 types of multi-task problems:reward-related multi-task,arising from different reward functions;multi-domain multi-task,caused by differences in state and action spaces,state transition functions;and scalability-related multi-task,resulting from the dynamic variation in the number of agents.Most existing studies focus on scalability-related multitask problems.However,with the increasing integration between large language models(LLMs)and multi-agent systems,a growing number of LLM-based multi-agent systems have emerged,enabling more complex multi-task cooperation.This paper provides a comprehensive review of the latest advances in this field.By combining multi-task reinforcement learning with cooperative MARL,we categorize and analyze the 3 major types of multi-task problems under multi-agent settings,offering more fine-grained classifications and summarizing key insights for each.In addition,we summarize commonly used benchmarks and discuss future directions of research in this area,which hold promise for further enhancing the multi-task cooperation capabilities of multi-agent systems and expanding their practical applications in the real world.展开更多
Predicting the health status of stroke patients at different stages of the disease is a critical clinical task.The onset and development of stroke are affected by an array of factors,encompassing genetic predispositio...Predicting the health status of stroke patients at different stages of the disease is a critical clinical task.The onset and development of stroke are affected by an array of factors,encompassing genetic predisposition,environmental exposure,unhealthy lifestyle habits,and existing medical conditions.Although existing machine learning-based methods for predicting stroke patients’health status have made significant progress,limitations remain in terms of prediction accuracy,model explainability,and system optimization.This paper proposes a multi-task learning approach based on Explainable Artificial Intelligence(XAI)for predicting the health status of stroke patients.First,we design a comprehensive multi-task learning framework that utilizes the task correlation of predicting various health status indicators in patients,enabling the parallel prediction of multiple health indicators.Second,we develop a multi-task Area Under Curve(AUC)optimization algorithm based on adaptive low-rank representation,which removes irrelevant information from the model structure to enhance the performance of multi-task AUC optimization.Additionally,the model’s explainability is analyzed through the stability analysis of SHAP values.Experimental results demonstrate that our approach outperforms comparison algorithms in key prognostic metrics F1 score and Efficiency.展开更多
Tropical cyclones(TCs)are one of the most serious types of natural disasters,and accurate TC activity predictions are key to disaster prevention and mitigation.Recently,TC track predictions have made significant progr...Tropical cyclones(TCs)are one of the most serious types of natural disasters,and accurate TC activity predictions are key to disaster prevention and mitigation.Recently,TC track predictions have made significant progress,but the ability to predict their intensity is obviously lagging behind.At present,research on TC intensity prediction takes atmospheric reanalysis data as the research object and mines the relationship between TC-related environmental factors and intensity through deep learning.However,reanalysis data are non-real-time in nature,which does not meet the requirements for operational forecasting applications.Therefore,a TC intensity prediction model named TC-Rolling is proposed,which can simultaneously extract the degree of symmetry for strong TC convective cloud and convection intensity,and fuse the deviation-angle variance with satellite images to construct the correlation between TC convection structure and intensity.For TCs'complex dynamic processes,a convolutional neural network(CNN)is used to learn their temporal and spatial features.For real-time intensity estimation,multi-task learning acts as an implicit time-series enhancement.The model is designed with a rolling strategy that aims to moderate the long-term dependent decay problem and improve accuracy for short-term intensity predictions.Since multiple tasks are correlated,the loss function of 12 h and 24 h are corrected.After testing on a sample of TCs in the Northwest Pacific,with a 4.48 kt root-mean-square error(RMSE)of 6 h intensity prediction,5.78 kt for 12 h,and 13.94 kt for 24 h,TC records from official agencies are used to assess the validity of TC-Rolling.展开更多
As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as...As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as a new challenge in the field of recommendation systems.This paper introduces a group-buying recommendation model based on multi-head attention mechanisms and multi-task learning,termed the Multi-head Attention Mechanisms and Multi-task Learning Group-Buying Recommendation(MAMGBR)model,specifically designed to optimize group-buying recommendations on e-commerce platforms.The core dataset of this study comes from the Chinese maternal and infant e-commerce platform“Beibei,”encompassing approximately 430,000 successful groupbuying actions and over 120,000 users.Themodel focuses on twomain tasks:recommending items for group organizers(Task Ⅰ)and recommending participants for a given group-buying event(Task Ⅱ).In model evaluation,MAMGBR achieves an MRR@10 of 0.7696 for Task I,marking a 20.23%improvement over baseline models.Furthermore,in Task II,where complex interaction patterns prevail,MAMGBR utilizes auxiliary loss functions to effectively model the multifaceted roles of users,items,and participants,leading to a 24.08%increase in MRR@100 under a 1:99 sample ratio.Experimental results show that compared to benchmark models,such as NGCF and EATNN,MAMGBR’s integration ofmulti-head attentionmechanisms,expert networks,and gating mechanisms enables more accurate modeling of user preferences and social associations within group-buying scenarios,significantly enhancing recommendation accuracy and platform group-buying success rates.展开更多
Aerosol optical depth(AOD)and fine particulate matter with a diameter of less than or equal to 2.5μm(PM_(2.5))play crucial roles in air quality,human health,and climate change.However,the complex correlation of AOD–...Aerosol optical depth(AOD)and fine particulate matter with a diameter of less than or equal to 2.5μm(PM_(2.5))play crucial roles in air quality,human health,and climate change.However,the complex correlation of AOD–PM_(2.5)and the limitations of existing algorithms pose a significant challenge in realizing the accurate joint retrieval of these two parameters at the same location.On this point,a multi-task learning(MTL)model,which enables the joint retrieval of PM_(2.5)concentration and AOD,is proposed and applied on the top-of-the-atmosphere reflectance data gathered by the Fengyun-4A Advanced Geosynchronous Radiation Imager(FY-4A AGRI),and compared to that of two single-task learning models—namely,Random Forest(RF)and Deep Neural Network(DNN).Specifically,MTL achieves a coefficient of determination(R^(2))of 0.88 and a root-mean-square error(RMSE)of 0.10 in AOD retrieval.In comparison to RF,the R^(2)increases by 0.04,the RMSE decreases by 0.02,and the percentage of retrieval results falling within the expected error range(Within-EE)rises by 5.55%.The R^(2)and RMSE of PM_(2.5)retrieval by MTL are 0.84 and 13.76μg m~(-3)respectively.Compared with RF,the R^(2)increases by 0.06,the RMSE decreases by 4.55μg m~(-3),and the Within-EE increases by 7.28%.Additionally,compared to DNN,MTL shows an increase of 0.01 in R^(2)and a decrease of 0.02 in RMSE in AOD retrieval,with a corresponding increase of 2.89%in Within-EE.For PM_(2.5)retrieval,MTL exhibits an increase of 0.05 in R^(2),a decrease of 1.76μg m~(-3)in RMSE,and an increase of 6.83%in Within-EE.The evaluation suggests that MTL is able to provide simultaneously improved AOD and PM_(2.5)retrievals,demonstrating a significant advantage in efficiently capturing the spatial distribution of PM_(2.5)concentration and AOD.展开更多
Accurate short-term forecast of offshore wind fields is still challenging for numerical weather prediction models.Based on three years of 48-hour forecast data from the European Centre for Medium-Range Weather Forecas...Accurate short-term forecast of offshore wind fields is still challenging for numerical weather prediction models.Based on three years of 48-hour forecast data from the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System global model(ECMWF-IFS)over 14 offshore weather stations along the coast of Shandong Province,this study introduces a multi-task learning(MTL)model(TabNet-MTL),which significantly improves the forecast bias of near-surface wind direction and speed simultaneously.TabNet-MTL adopts the feature engineering method,utilizes mean square error as the loss function,and employs the 5-fold cross validation method to ensure the generalization ability of the trained model.It demonstrates superior skills in wind field correction across different forecast lead times over all stations compared to its single-task version(TabNet-STL)and three other popular single-task learning models(Random Forest,LightGBM,and XGBoost).Results show that it significantly reduces root mean square error of the ECMWF-IFS wind speed forecast from 2.20 to 1.25 m s−1,and increases the forecast accuracy of wind direction from 50%to 65%.As an explainable deep learning model,the weather stations and long-term temporal statistics of near-surface wind speed are identified as the most influential variables for TabNet-MTL in constructing its feature engineering.展开更多
The livestock farming is an important pillar of the rural economy in China.To explore the impact of government technical subsidies and pollution penalties on the digital and intelligent transformation of livestock ent...The livestock farming is an important pillar of the rural economy in China.To explore the impact of government technical subsidies and pollution penalties on the digital and intelligent transformation of livestock enterprises,an evolutionary game theoretical model between the government and livestock enterprises is constructed.The interaction mechanism of the game between the government and breeding enterprises is explored,and simulation is conducted.The research results show that the combined strategy of pollution penalties and technical subsidies is the optimal strategy for the government;the system is jointly driven by government subsidies,technical costs of transformation input,public willingness,and enterprise willingness.展开更多
The accurate prediction of drug absorption,distribution,metabolism,excretion,and toxicity(ADMET)properties represents a crucial step in early drug development for reducing failure risk.Current deep learning approaches...The accurate prediction of drug absorption,distribution,metabolism,excretion,and toxicity(ADMET)properties represents a crucial step in early drug development for reducing failure risk.Current deep learning approaches face challenges with data sparsity and information loss due to single-molecule representation limitations and isolated predictive tasks.This research proposes molecular properties prediction with parallel-view and collaborative learning(MolP-PC),a multi-view fusion and multi-task deep learning framework that integrates 1D molecular fingerprints(MFs),2D molecular graphs,and 3D geometric representations,incorporating an attention-gated fusion mechanism and multi-task adaptive learning strategy for precise ADMET property predictions.Experimental results demonstrate that MolP-PC achieves optimal performance in 27 of 54 tasks,with its multi-task learning(MTL)mechanism significantly enhancing predictive performance on small-scale datasets and surpassing single-task models in 41 of 54 tasks.Additional ablation studies and interpretability analyses confirm the significance of multi-view fusion in capturing multi-dimensional molecular information and enhancing model generalization.A case study examining the anticancer compound Oroxylin A demonstrates MolP-PC’s effective generalization in predicting key pharmacokinetic parameters such as half-life(T0.5)and clearance(CL),indicating its practical utility in drug modeling.However,the model exhibits a tendency to underestimate volume of distribution(VD),indicating potential for improvement in analyzing compounds with high tissue distribution.This study presents an efficient and interpretable approach for ADMET property prediction,establishing a novel framework for molecular optimization and risk assessment in drug development.展开更多
As a key component of the plant antioxidant enzymatic system,superoxide dismutase(SOD)can efficiently protect cells from oxidative stress and maintain redox homeostasis.Currently,there are few studies related to SOD g...As a key component of the plant antioxidant enzymatic system,superoxide dismutase(SOD)can efficiently protect cells from oxidative stress and maintain redox homeostasis.Currently,there are few studies related to SOD genes in various taxa of algae,and the specific functions and evolutionary patterns of these family members remain unclear.In this study,comprehensively evolutionary analysis of SOD gene family in the bladed Bangiales was carried out.A total of 9,10,and 12 SOD genes were identified from three species of Pophyra umbilicalis,Pyropia haitanensis,and Pyropia yezoensis,respectively.Based on phylogenetic analysis,SOD gene members within the same subfamily exhibited similar motif patterns as well as conserved domains,which could be attribute to Cu/Zn-SOD and Fe/Mn-SOD.The promoter regions of SOD genes were rich in hormone-responsive,stress-responsive,and growth cis-acting elements,with variations and similarities observed among different species of other red algae and subfamilies.According to subcellular location prediction,it is suggested that Cu/Zn-SOD was predominantly located in chloroplasts,while Fe/Mn-SOD was primarily located in mitochondria.Also,the two subfamilies differed significantly in the two-/three-dimensional protein structures.In terms of gene evolution,the strongest collinearity relationship was shown between Pyropia haitanensis and Pyropia yezoensis,with all the 1꞉1 orthologous gene pair being subjected to a purifying selection(Ka/Ks<1,Ka:non-synonymy rate;Ks:synonymy rate).Moreover,12 SOD genes underwent positive selection during the evolutionary process.Furthermore,gene expression analysis based on transcriptomic data from Pyropia haitanensis showed that the expression patterns of SOD genes varied under different stress conditions.Together,this study revealed the evolutionary pattern of SOD genes in three bladed Bangiales species,which will lay the foundation for subsequent studies on the function of SOD genes.展开更多
How ecological and evolutionary factors affect small mammal diversity in arid regions remains largely unknown.Here,we combined the largest phylogeny and occurrence dataset of Gerbillinae desert rodents to explore the ...How ecological and evolutionary factors affect small mammal diversity in arid regions remains largely unknown.Here,we combined the largest phylogeny and occurrence dataset of Gerbillinae desert rodents to explore the underlying factors shaping present-day distribution patterns.In particular,we analyzed the relative contributions of ecological and evolutionary factors on their species diversity using a variety of models.Additionally,we inferred the ancestral range and possible dispersal scenarios and estimated the diversification rate of Gerbillinae.We found that Gerbillinae likely originated in the Horn of Africa in the Middle Miocene and then dispersed and diversified across arid regions in northern and southern Africa and western and central Asia,forming their current distribution pattern.Multiple ecological and evolutionary factors jointly determine the spatial pattern of Gerbillinae diversity,but evolutionary factors(evolutionary time and speciation rate)and habitat filtering were the most important in explaining the spatial variation in species richness.Our study enhances the understanding of the diversity patterns of small mammals in arid regions and highlights the importance of including evolutionary factors when interpreting the mechanisms underlying large-scale species diversity patterns.展开更多
In recent years,the development of new types of nuclear reactors,such as transportable,marine,and space reactors,has presented new challenges for the optimization of reactor radiation-shielding design.Shielding struct...In recent years,the development of new types of nuclear reactors,such as transportable,marine,and space reactors,has presented new challenges for the optimization of reactor radiation-shielding design.Shielding structures typically need to be lightweight,miniaturized,and radiation-protected,which is a multi-parameter and multi-objective optimization problem.The conventional multi-objective(two or three objectives)optimization method for radiation-shielding design exhibits limitations for a number of optimization objectives and variable parameters,as well as a deficiency in achieving a global optimal solution,thereby failing to meet the requirements of shielding optimization for newly developed reactors.In this study,genetic and artificial bee-colony algorithms are combined with a reference-point-selection strategy and applied to the many-objective(having four or more objectives)optimal design of reactor radiation shielding.To validate the reliability of the methods,an optimization simulation is conducted on three-dimensional shielding structures and another complicated shielding-optimization problem.The numerical results demonstrate that the proposed algorithms outperform conventional shielding-design methods in terms of optimization performance,and they exhibit their reliability in practical engineering problems.The many-objective optimization algorithms developed in this study are proven to efficiently and consistently search for Pareto-front shielding schemes.Therefore,the algorithms proposed in this study offer novel insights into improving the shielding-design performance and shielding quality of new reactor types.展开更多
The yellow boxfish(Ostracion cubicus)exhibits a combination of derived morphological traits specialized for coral reef environments and ancestral characteristics,including a fused dermal plate.Contradictory evolutiona...The yellow boxfish(Ostracion cubicus)exhibits a combination of derived morphological traits specialized for coral reef environments and ancestral characteristics,including a fused dermal plate.Contradictory evolutionary evidence hinders true classification of O.cubicus.To clarify its evolutionary position within Tetraodontiformes,a chromosome-level genome assembly was generated,representing the most contiguous and complete genome to date for this lineage.Notably,O.cubicus possessed the largest genome within the order Tetraodontiformes,primarily due to extensive transposable element expansion.Phylogenetic analysis based on 19 whole genomes and 131 mitochondrial genomes resolved Tetraodontiformes into three major sister groups(Ostraciidae-Molidae,Tetraodontidae,and Balistidae-Monacanthidae).Comparative genomic evidence indicated that O.cubicus diverged early from the common ancestor of modern Tetraodontiformes and retained the highest number of HOX genes among surveyed taxa.Although overall genomic architecture was largely conserved,certain genetic and environmental changes may have contributed to its phenotypic adaptations,including climate cooling during the Miocene-Pliocene Transition,recent DNA and long interspersed nuclear element(LINE)transposon bursts,lineage-specific chromosomal rearrangements,and gene family expansion.Many positively selected genes and rapidly evolving genes were associated with skeletal development,including bmp7,egf7,and bmpr2.Transcriptomic comparisons between carapace and tail skin revealed various candidate genes and pathways related to carapace formation,such as postn,scpp1,and components of the TGF-βsignaling pathway.A derived amino acid substitution in eda,coupled with protein structural modeling,suggested potential molecular convergence in dermal plate formation among teleosts.These findings provide novel insights into the genomic and developmental basis of carapace evolution and coral reef-adaptation in O.cubicus,offering a strong case for evolutionary balance between genomic conservation with regulatory innovation to achieve coral reef specialization.展开更多
Multi-firmware comparison techniques can improve efficiency when auditing firmwares in bulk.How-ever,the problem of matching functions between multiple firmwares has not been studied before.This paper proposes a multi...Multi-firmware comparison techniques can improve efficiency when auditing firmwares in bulk.How-ever,the problem of matching functions between multiple firmwares has not been studied before.This paper proposes a multi-firmware comparison method based on evolutionary algorithms and trusted base points.We first model the multi-firmware comparison as a multi-sequence matching problem.Then,we propose an adaptation function and a population generation method based on trusted base points.Finally,we apply an evolutionary algorithm to find the optimal result.At the same time,we design the similarity of matching results as an evaluation metric to measure the effect of multi-firmware comparison.The experiments show that the proposed method outperforms Bindiff and the string-based method.Precisely,the similarity between the matching results of the proposed method and Bindiff matching results is 61%,and the similarity between the matching results of the proposed method and the string-based method is 62.8%.By sampling and manual verification,the accuracy of the matching results of the proposed method can be about 66.4%.展开更多
Accurate forecasting of blast furnace gas(BFG)production is an essential prerequisite for reasonable energy scheduling and management to reduce carbon emissions.Coupling forecasting between BFG generation and consumpt...Accurate forecasting of blast furnace gas(BFG)production is an essential prerequisite for reasonable energy scheduling and management to reduce carbon emissions.Coupling forecasting between BFG generation and consumption dynamics was taken as the research object.A multi-task learning(MTL)method for BFG forecasting was proposed,which integrated a coupling correlation coefficient(CCC)and an inverted transformer structure.The CCC method could enhance key information extraction by establishing relationships between multiple prediction targets and relevant factors,while MTL effectively captured the inherent correlations between BFG generation and consumption.Finally,a real-world case study was conducted to compare the proposed model with four benchmark models.Results indicated significant reductions in average mean absolute percentage error by 33.37%,achieving 1.92%,with a computational time of 76 s.The sensitivity analysis of hyperparameters such as learning rate,batch size,and units of the long short-term memory layer highlights the importance of hyperparameter tuning.展开更多
In this paper,a robust and consistent COVID-19 emergency decision-making approach is proposed based on q-rung linear diophantine fuzzy set(q-RLDFS),differential evolutionary(DE)optimization principles,and evidential r...In this paper,a robust and consistent COVID-19 emergency decision-making approach is proposed based on q-rung linear diophantine fuzzy set(q-RLDFS),differential evolutionary(DE)optimization principles,and evidential reasoning(ER)methodology.The proposed approach uses q-RLDFS in order to represent the evaluating values of the alternatives corresponding to the attributes.DE optimization is used to obtain the optimal weights of the attributes,and ER methodology is used to compute the aggregated q-rung linear diophantine fuzzy values(q-RLDFVs)of each alternative.Then the score values of alternatives are computed based on the aggregated q-RLDFVs.An alternative with the maximum score value is selected as a better one.The applicability of the proposed approach has been illustrated in COVID-19 emergency decision-making system and sustainable energy planning management.Moreover,we have validated the proposed approach with a numerical example.Finally,a comparative study is provided with the existing models,where the proposed approach is found to be robust to perform better and consistent in uncertain environments.展开更多
基金supported by the Open Project of Xiangjiang Laboratory(No.22XJ02003)the National Natural Science Foundation of China(No.62122093).
文摘Time series clustering is a challenging problem due to the large-volume,high-dimensional,and warping characteristics of time series data.Traditional clustering methods often use a single criterion or distance measure,which may not capture all the features of the data.This paper proposes a novel method for time series clustering based on evolutionary multi-tasking optimization,termed i-MFEA,which uses an improved multifactorial evolutionary algorithm to optimize multiple clustering tasks simultaneously,each with a different validity index or distance measure.Therefore,i-MFEA can produce diverse and robust clustering solutions that satisfy various preferences of decision-makers.Experiments on two artificial datasets show that i-MFEA outperforms single-objective evolutionary algorithms and traditional clustering methods in terms of convergence speed and clustering quality.The paper also discusses how i-MFEA can address two long-standing issues in time series clustering:the choice of appropriate similarity measure and the number of clusters.
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.
基金The National Natural Science Foundation of China(62136008,62293541)The Beijing Natural Science Foundation(4232056)The Beijing Nova Program(20240484514).
文摘Cooperative multi-agent reinforcement learning(MARL)is a key technology for enabling cooperation in complex multi-agent systems.It has achieved remarkable progress in areas such as gaming,autonomous driving,and multi-robot control.Empowering cooperative MARL with multi-task decision-making capabilities is expected to further broaden its application scope.In multi-task scenarios,cooperative MARL algorithms need to address 3 types of multi-task problems:reward-related multi-task,arising from different reward functions;multi-domain multi-task,caused by differences in state and action spaces,state transition functions;and scalability-related multi-task,resulting from the dynamic variation in the number of agents.Most existing studies focus on scalability-related multitask problems.However,with the increasing integration between large language models(LLMs)and multi-agent systems,a growing number of LLM-based multi-agent systems have emerged,enabling more complex multi-task cooperation.This paper provides a comprehensive review of the latest advances in this field.By combining multi-task reinforcement learning with cooperative MARL,we categorize and analyze the 3 major types of multi-task problems under multi-agent settings,offering more fine-grained classifications and summarizing key insights for each.In addition,we summarize commonly used benchmarks and discuss future directions of research in this area,which hold promise for further enhancing the multi-task cooperation capabilities of multi-agent systems and expanding their practical applications in the real world.
基金funded by the Excellent Talent Training Funding Project in Dongcheng District,Beijing,with project number 2024-dchrcpyzz-9.
文摘Predicting the health status of stroke patients at different stages of the disease is a critical clinical task.The onset and development of stroke are affected by an array of factors,encompassing genetic predisposition,environmental exposure,unhealthy lifestyle habits,and existing medical conditions.Although existing machine learning-based methods for predicting stroke patients’health status have made significant progress,limitations remain in terms of prediction accuracy,model explainability,and system optimization.This paper proposes a multi-task learning approach based on Explainable Artificial Intelligence(XAI)for predicting the health status of stroke patients.First,we design a comprehensive multi-task learning framework that utilizes the task correlation of predicting various health status indicators in patients,enabling the parallel prediction of multiple health indicators.Second,we develop a multi-task Area Under Curve(AUC)optimization algorithm based on adaptive low-rank representation,which removes irrelevant information from the model structure to enhance the performance of multi-task AUC optimization.Additionally,the model’s explainability is analyzed through the stability analysis of SHAP values.Experimental results demonstrate that our approach outperforms comparison algorithms in key prognostic metrics F1 score and Efficiency.
基金jointly supported by the National Natural Science Foundation of China(Grant Nos.42075138 and 42375147)the Program on Key Basic Research Project of Jiangsu(Grant No.BE2023829)。
文摘Tropical cyclones(TCs)are one of the most serious types of natural disasters,and accurate TC activity predictions are key to disaster prevention and mitigation.Recently,TC track predictions have made significant progress,but the ability to predict their intensity is obviously lagging behind.At present,research on TC intensity prediction takes atmospheric reanalysis data as the research object and mines the relationship between TC-related environmental factors and intensity through deep learning.However,reanalysis data are non-real-time in nature,which does not meet the requirements for operational forecasting applications.Therefore,a TC intensity prediction model named TC-Rolling is proposed,which can simultaneously extract the degree of symmetry for strong TC convective cloud and convection intensity,and fuse the deviation-angle variance with satellite images to construct the correlation between TC convection structure and intensity.For TCs'complex dynamic processes,a convolutional neural network(CNN)is used to learn their temporal and spatial features.For real-time intensity estimation,multi-task learning acts as an implicit time-series enhancement.The model is designed with a rolling strategy that aims to moderate the long-term dependent decay problem and improve accuracy for short-term intensity predictions.Since multiple tasks are correlated,the loss function of 12 h and 24 h are corrected.After testing on a sample of TCs in the Northwest Pacific,with a 4.48 kt root-mean-square error(RMSE)of 6 h intensity prediction,5.78 kt for 12 h,and 13.94 kt for 24 h,TC records from official agencies are used to assess the validity of TC-Rolling.
基金supported by the Key Research and Development Program of Heilongjiang Province(No.2022ZX01A35).
文摘As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as a new challenge in the field of recommendation systems.This paper introduces a group-buying recommendation model based on multi-head attention mechanisms and multi-task learning,termed the Multi-head Attention Mechanisms and Multi-task Learning Group-Buying Recommendation(MAMGBR)model,specifically designed to optimize group-buying recommendations on e-commerce platforms.The core dataset of this study comes from the Chinese maternal and infant e-commerce platform“Beibei,”encompassing approximately 430,000 successful groupbuying actions and over 120,000 users.Themodel focuses on twomain tasks:recommending items for group organizers(Task Ⅰ)and recommending participants for a given group-buying event(Task Ⅱ).In model evaluation,MAMGBR achieves an MRR@10 of 0.7696 for Task I,marking a 20.23%improvement over baseline models.Furthermore,in Task II,where complex interaction patterns prevail,MAMGBR utilizes auxiliary loss functions to effectively model the multifaceted roles of users,items,and participants,leading to a 24.08%increase in MRR@100 under a 1:99 sample ratio.Experimental results show that compared to benchmark models,such as NGCF and EATNN,MAMGBR’s integration ofmulti-head attentionmechanisms,expert networks,and gating mechanisms enables more accurate modeling of user preferences and social associations within group-buying scenarios,significantly enhancing recommendation accuracy and platform group-buying success rates.
基金supported by the National Natural Science Foundation of China(Grant Nos.42030708,42375138,42030608,42105128,42075079)the Opening Foundation of Key Laboratory of Atmospheric Sounding,China Meteorological Administration(CMA),and the CMA Research Center on Meteorological Observation Engineering Technology(Grant No.U2021Z03),and the Opening Foundation of the Key Laboratory of Atmospheric Chemistry,CMA(Grant No.2022B02)。
文摘Aerosol optical depth(AOD)and fine particulate matter with a diameter of less than or equal to 2.5μm(PM_(2.5))play crucial roles in air quality,human health,and climate change.However,the complex correlation of AOD–PM_(2.5)and the limitations of existing algorithms pose a significant challenge in realizing the accurate joint retrieval of these two parameters at the same location.On this point,a multi-task learning(MTL)model,which enables the joint retrieval of PM_(2.5)concentration and AOD,is proposed and applied on the top-of-the-atmosphere reflectance data gathered by the Fengyun-4A Advanced Geosynchronous Radiation Imager(FY-4A AGRI),and compared to that of two single-task learning models—namely,Random Forest(RF)and Deep Neural Network(DNN).Specifically,MTL achieves a coefficient of determination(R^(2))of 0.88 and a root-mean-square error(RMSE)of 0.10 in AOD retrieval.In comparison to RF,the R^(2)increases by 0.04,the RMSE decreases by 0.02,and the percentage of retrieval results falling within the expected error range(Within-EE)rises by 5.55%.The R^(2)and RMSE of PM_(2.5)retrieval by MTL are 0.84 and 13.76μg m~(-3)respectively.Compared with RF,the R^(2)increases by 0.06,the RMSE decreases by 4.55μg m~(-3),and the Within-EE increases by 7.28%.Additionally,compared to DNN,MTL shows an increase of 0.01 in R^(2)and a decrease of 0.02 in RMSE in AOD retrieval,with a corresponding increase of 2.89%in Within-EE.For PM_(2.5)retrieval,MTL exhibits an increase of 0.05 in R^(2),a decrease of 1.76μg m~(-3)in RMSE,and an increase of 6.83%in Within-EE.The evaluation suggests that MTL is able to provide simultaneously improved AOD and PM_(2.5)retrievals,demonstrating a significant advantage in efficiently capturing the spatial distribution of PM_(2.5)concentration and AOD.
基金the National Key Research and Development Plan of China[Grant No.2023YFB3002400]the Shanghai 2021 Natural Science Foundation[Grant Nos.21ZR1420400 and 21ZR1419800]+1 种基金the Shanghai 2023 Natural Science Foundation[Grant No.23ZR1463000]the Shandong Provincial Meteorological Bureau Scientific Research Project[Grant No.2023SDBD05].
文摘Accurate short-term forecast of offshore wind fields is still challenging for numerical weather prediction models.Based on three years of 48-hour forecast data from the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System global model(ECMWF-IFS)over 14 offshore weather stations along the coast of Shandong Province,this study introduces a multi-task learning(MTL)model(TabNet-MTL),which significantly improves the forecast bias of near-surface wind direction and speed simultaneously.TabNet-MTL adopts the feature engineering method,utilizes mean square error as the loss function,and employs the 5-fold cross validation method to ensure the generalization ability of the trained model.It demonstrates superior skills in wind field correction across different forecast lead times over all stations compared to its single-task version(TabNet-STL)and three other popular single-task learning models(Random Forest,LightGBM,and XGBoost).Results show that it significantly reduces root mean square error of the ECMWF-IFS wind speed forecast from 2.20 to 1.25 m s−1,and increases the forecast accuracy of wind direction from 50%to 65%.As an explainable deep learning model,the weather stations and long-term temporal statistics of near-surface wind speed are identified as the most influential variables for TabNet-MTL in constructing its feature engineering.
文摘The livestock farming is an important pillar of the rural economy in China.To explore the impact of government technical subsidies and pollution penalties on the digital and intelligent transformation of livestock enterprises,an evolutionary game theoretical model between the government and livestock enterprises is constructed.The interaction mechanism of the game between the government and breeding enterprises is explored,and simulation is conducted.The research results show that the combined strategy of pollution penalties and technical subsidies is the optimal strategy for the government;the system is jointly driven by government subsidies,technical costs of transformation input,public willingness,and enterprise willingness.
基金supported by the research on key technologies for monitoring and identifying drug abuse of anesthetic drugs and psychotropic drugs,and intervention for addiction(No.2023YFC3304200)the program of a study on the diagnosis of addiction to synthetic cannabinoids and methods of assessing the risk of abuse(No.2022YFC3300905)+1 种基金the program of Ab initio design and generation of AI models for small molecule ligands based on target structures(No.2022PE0AC03)ZHIJIANG LAB.
文摘The accurate prediction of drug absorption,distribution,metabolism,excretion,and toxicity(ADMET)properties represents a crucial step in early drug development for reducing failure risk.Current deep learning approaches face challenges with data sparsity and information loss due to single-molecule representation limitations and isolated predictive tasks.This research proposes molecular properties prediction with parallel-view and collaborative learning(MolP-PC),a multi-view fusion and multi-task deep learning framework that integrates 1D molecular fingerprints(MFs),2D molecular graphs,and 3D geometric representations,incorporating an attention-gated fusion mechanism and multi-task adaptive learning strategy for precise ADMET property predictions.Experimental results demonstrate that MolP-PC achieves optimal performance in 27 of 54 tasks,with its multi-task learning(MTL)mechanism significantly enhancing predictive performance on small-scale datasets and surpassing single-task models in 41 of 54 tasks.Additional ablation studies and interpretability analyses confirm the significance of multi-view fusion in capturing multi-dimensional molecular information and enhancing model generalization.A case study examining the anticancer compound Oroxylin A demonstrates MolP-PC’s effective generalization in predicting key pharmacokinetic parameters such as half-life(T0.5)and clearance(CL),indicating its practical utility in drug modeling.However,the model exhibits a tendency to underestimate volume of distribution(VD),indicating potential for improvement in analyzing compounds with high tissue distribution.This study presents an efficient and interpretable approach for ADMET property prediction,establishing a novel framework for molecular optimization and risk assessment in drug development.
基金Supported by the National Key R&D Program of China(No.2023 YFD 2400102)the Guangxi Key Laboratory of Beibu Gulf Marine Biodiversity Conservation,Beibu Gulf University(No.2024 KA 04)。
文摘As a key component of the plant antioxidant enzymatic system,superoxide dismutase(SOD)can efficiently protect cells from oxidative stress and maintain redox homeostasis.Currently,there are few studies related to SOD genes in various taxa of algae,and the specific functions and evolutionary patterns of these family members remain unclear.In this study,comprehensively evolutionary analysis of SOD gene family in the bladed Bangiales was carried out.A total of 9,10,and 12 SOD genes were identified from three species of Pophyra umbilicalis,Pyropia haitanensis,and Pyropia yezoensis,respectively.Based on phylogenetic analysis,SOD gene members within the same subfamily exhibited similar motif patterns as well as conserved domains,which could be attribute to Cu/Zn-SOD and Fe/Mn-SOD.The promoter regions of SOD genes were rich in hormone-responsive,stress-responsive,and growth cis-acting elements,with variations and similarities observed among different species of other red algae and subfamilies.According to subcellular location prediction,it is suggested that Cu/Zn-SOD was predominantly located in chloroplasts,while Fe/Mn-SOD was primarily located in mitochondria.Also,the two subfamilies differed significantly in the two-/three-dimensional protein structures.In terms of gene evolution,the strongest collinearity relationship was shown between Pyropia haitanensis and Pyropia yezoensis,with all the 1꞉1 orthologous gene pair being subjected to a purifying selection(Ka/Ks<1,Ka:non-synonymy rate;Ks:synonymy rate).Moreover,12 SOD genes underwent positive selection during the evolutionary process.Furthermore,gene expression analysis based on transcriptomic data from Pyropia haitanensis showed that the expression patterns of SOD genes varied under different stress conditions.Together,this study revealed the evolutionary pattern of SOD genes in three bladed Bangiales species,which will lay the foundation for subsequent studies on the function of SOD genes.
基金supported by grants from the Third Xinjiang Scientific Expedition Program(Grant No.2022xjkk0205 to Lin Xia,No.2021xjkk0604 to Jilong Cheng)the National Natural Science Foundation of China(32170416 to Qisen Yang,31900325 to Jilong Cheng)+1 种基金the Joint Fund of National Natural Science Foundation of China(U2003203 to Lin Xia)the Key Laboratory of Zoological Systematics and Evolution of the Chinese Academy of Sciences(Y229YX5105 to Qisen Yang).
文摘How ecological and evolutionary factors affect small mammal diversity in arid regions remains largely unknown.Here,we combined the largest phylogeny and occurrence dataset of Gerbillinae desert rodents to explore the underlying factors shaping present-day distribution patterns.In particular,we analyzed the relative contributions of ecological and evolutionary factors on their species diversity using a variety of models.Additionally,we inferred the ancestral range and possible dispersal scenarios and estimated the diversification rate of Gerbillinae.We found that Gerbillinae likely originated in the Horn of Africa in the Middle Miocene and then dispersed and diversified across arid regions in northern and southern Africa and western and central Asia,forming their current distribution pattern.Multiple ecological and evolutionary factors jointly determine the spatial pattern of Gerbillinae diversity,but evolutionary factors(evolutionary time and speciation rate)and habitat filtering were the most important in explaining the spatial variation in species richness.Our study enhances the understanding of the diversity patterns of small mammals in arid regions and highlights the importance of including evolutionary factors when interpreting the mechanisms underlying large-scale species diversity patterns.
基金supported by the National Natural Science Foundation of China(Nos.12475174 and 12175101)Yue Lu Shan Center Industrial Innovation(No.2024YCII0108)。
文摘In recent years,the development of new types of nuclear reactors,such as transportable,marine,and space reactors,has presented new challenges for the optimization of reactor radiation-shielding design.Shielding structures typically need to be lightweight,miniaturized,and radiation-protected,which is a multi-parameter and multi-objective optimization problem.The conventional multi-objective(two or three objectives)optimization method for radiation-shielding design exhibits limitations for a number of optimization objectives and variable parameters,as well as a deficiency in achieving a global optimal solution,thereby failing to meet the requirements of shielding optimization for newly developed reactors.In this study,genetic and artificial bee-colony algorithms are combined with a reference-point-selection strategy and applied to the many-objective(having four or more objectives)optimal design of reactor radiation shielding.To validate the reliability of the methods,an optimization simulation is conducted on three-dimensional shielding structures and another complicated shielding-optimization problem.The numerical results demonstrate that the proposed algorithms outperform conventional shielding-design methods in terms of optimization performance,and they exhibit their reliability in practical engineering problems.The many-objective optimization algorithms developed in this study are proven to efficiently and consistently search for Pareto-front shielding schemes.Therefore,the algorithms proposed in this study offer novel insights into improving the shielding-design performance and shielding quality of new reactor types.
基金supported by the Basic and Applied Basic Research Foundation of Guangdong Province(2020A1515110882)National Science Fund for Distinguished Young Scholars(32225049)。
文摘The yellow boxfish(Ostracion cubicus)exhibits a combination of derived morphological traits specialized for coral reef environments and ancestral characteristics,including a fused dermal plate.Contradictory evolutionary evidence hinders true classification of O.cubicus.To clarify its evolutionary position within Tetraodontiformes,a chromosome-level genome assembly was generated,representing the most contiguous and complete genome to date for this lineage.Notably,O.cubicus possessed the largest genome within the order Tetraodontiformes,primarily due to extensive transposable element expansion.Phylogenetic analysis based on 19 whole genomes and 131 mitochondrial genomes resolved Tetraodontiformes into three major sister groups(Ostraciidae-Molidae,Tetraodontidae,and Balistidae-Monacanthidae).Comparative genomic evidence indicated that O.cubicus diverged early from the common ancestor of modern Tetraodontiformes and retained the highest number of HOX genes among surveyed taxa.Although overall genomic architecture was largely conserved,certain genetic and environmental changes may have contributed to its phenotypic adaptations,including climate cooling during the Miocene-Pliocene Transition,recent DNA and long interspersed nuclear element(LINE)transposon bursts,lineage-specific chromosomal rearrangements,and gene family expansion.Many positively selected genes and rapidly evolving genes were associated with skeletal development,including bmp7,egf7,and bmpr2.Transcriptomic comparisons between carapace and tail skin revealed various candidate genes and pathways related to carapace formation,such as postn,scpp1,and components of the TGF-βsignaling pathway.A derived amino acid substitution in eda,coupled with protein structural modeling,suggested potential molecular convergence in dermal plate formation among teleosts.These findings provide novel insights into the genomic and developmental basis of carapace evolution and coral reef-adaptation in O.cubicus,offering a strong case for evolutionary balance between genomic conservation with regulatory innovation to achieve coral reef specialization.
文摘Multi-firmware comparison techniques can improve efficiency when auditing firmwares in bulk.How-ever,the problem of matching functions between multiple firmwares has not been studied before.This paper proposes a multi-firmware comparison method based on evolutionary algorithms and trusted base points.We first model the multi-firmware comparison as a multi-sequence matching problem.Then,we propose an adaptation function and a population generation method based on trusted base points.Finally,we apply an evolutionary algorithm to find the optimal result.At the same time,we design the similarity of matching results as an evaluation metric to measure the effect of multi-firmware comparison.The experiments show that the proposed method outperforms Bindiff and the string-based method.Precisely,the similarity between the matching results of the proposed method and Bindiff matching results is 61%,and the similarity between the matching results of the proposed method and the string-based method is 62.8%.By sampling and manual verification,the accuracy of the matching results of the proposed method can be about 66.4%.
基金supported by the National Natural Science Foundation of China(No.52474435)China Baowu Low Carbon Metallurgy Innovation Foundation(BWLCF202307).
文摘Accurate forecasting of blast furnace gas(BFG)production is an essential prerequisite for reasonable energy scheduling and management to reduce carbon emissions.Coupling forecasting between BFG generation and consumption dynamics was taken as the research object.A multi-task learning(MTL)method for BFG forecasting was proposed,which integrated a coupling correlation coefficient(CCC)and an inverted transformer structure.The CCC method could enhance key information extraction by establishing relationships between multiple prediction targets and relevant factors,while MTL effectively captured the inherent correlations between BFG generation and consumption.Finally,a real-world case study was conducted to compare the proposed model with four benchmark models.Results indicated significant reductions in average mean absolute percentage error by 33.37%,achieving 1.92%,with a computational time of 76 s.The sensitivity analysis of hyperparameters such as learning rate,batch size,and units of the long short-term memory layer highlights the importance of hyperparameter tuning.
文摘In this paper,a robust and consistent COVID-19 emergency decision-making approach is proposed based on q-rung linear diophantine fuzzy set(q-RLDFS),differential evolutionary(DE)optimization principles,and evidential reasoning(ER)methodology.The proposed approach uses q-RLDFS in order to represent the evaluating values of the alternatives corresponding to the attributes.DE optimization is used to obtain the optimal weights of the attributes,and ER methodology is used to compute the aggregated q-rung linear diophantine fuzzy values(q-RLDFVs)of each alternative.Then the score values of alternatives are computed based on the aggregated q-RLDFVs.An alternative with the maximum score value is selected as a better one.The applicability of the proposed approach has been illustrated in COVID-19 emergency decision-making system and sustainable energy planning management.Moreover,we have validated the proposed approach with a numerical example.Finally,a comparative study is provided with the existing models,where the proposed approach is found to be robust to perform better and consistent in uncertain environments.
