Surface-supported clusters forming by aggregation of excessive adatoms could be the main defects of 2D materials after chemical vapor deposition.They will significantly impact the electronic/magnetic properties.Moreov...Surface-supported clusters forming by aggregation of excessive adatoms could be the main defects of 2D materials after chemical vapor deposition.They will significantly impact the electronic/magnetic properties.Moreover,surface supported atoms are also widely explored for high active and selecting catalysts.Severe deformation,even dipping into the surface,of these clusters can be expected because of the very active edge of clusters and strong interaction between supported clusters and surfaces.However,most models of these clusters are supposed to simply float on the top of the surface because ab initio simulations cannot afford the complex reconstructions.Here,we develop an accurate graph neural network machine learning potential(MLP)from ab initio data by active learning architecture through fine-tuning pre-trained models,and then employ the MLP into Monte Carlo to explore the structural evolutions of Mo and S clusters(1-8 atoms)on perfect and various defective MoS2 monolayers.Interestingly,Mo clusters can always sink and embed themselves into MoS2 layers.In contrast,S clusters float on perfect surfaces.On the defective surface,a few S atoms will fill the vacancy and rest S clusters float on the top.Such significant structural reconstructions should be carefully taken into account.展开更多
In materials science,a significant correlation often exists between material input parameters and their corresponding performance attributes.Nevertheless,the inherent challenges associated with small data obscure thes...In materials science,a significant correlation often exists between material input parameters and their corresponding performance attributes.Nevertheless,the inherent challenges associated with small data obscure these statistical correlations,impeding machine learning models from effectively capturing the underlying patterns,thereby hampering efficient optimization of material properties.This work presents a novel active learning framework that integrates generative adversarial networks(GAN)with a directionally constrained expected absolute improvement(EAI)acquisition function to accelerate the discovery of ultra-high temperature ceramics(UHTCs)using small data.The framework employs GAN for data augmentation,symbolic regression for feature weight derivation,and a self-developed EAI function that incorporates input feature importance weighting to quantify bidirectional deviations from zero ablation rate.Through only two iterations,this framework successfully identified the optimal composition of HfB_(2)-3.52SiC-5.23TaSi_(2),which exhibits robust near-zero ablation rates under plasma ablation at 2500℃ for 200 s,demonstrating superior sampling efficiency compared to conventional active learning approaches.Microstructural analysis reveals that the exceptional performance stems from the formation of a highly viscous HfO_(2)-SiO_(2)-Ta_(2)O_(5)-HfSiO_(4)-Hf_(3)(BO_(3))_(4) oxide layer,which provides effective oxygen barrier protection.This work demonstrates an efficient and universal approach for rapid materials discovery using small data.展开更多
Industrial decarbonization is critical for achieving net-zero goals.The carbon dioxide electrochemical reduction reaction(CO_(2)RR)is a promising approach for converting CO_(2)into high-value chemicals,offering the po...Industrial decarbonization is critical for achieving net-zero goals.The carbon dioxide electrochemical reduction reaction(CO_(2)RR)is a promising approach for converting CO_(2)into high-value chemicals,offering the potential for decarbonizing industrial processes toward a sustainable,carbon-neutral future.However,developing CO_(2)RR catalysts with high selectivity and activity remains a challenge due to the complexity of finding such catalysts and the inefficiency of traditional computational or experimental approaches.Here,we present a methodology integrating density functional theory(DFT)calculations,deep learning models,and an active learning strategy to rapidly screen high-performance catalysts.The proposed methodology is then demonstrated on graphene-based single-atom catalysts for selective CO_(2)electroreduction to methanol.First,we conduct systematic binding energy calculations for 3045 single-atom catalysts to identify thermodynamically stable catalysts as the design space.We then use a graph neural network,fine-tuned with a specialized adsorption energy database,to predict the relative activity and selectivity of the candidate catalysts.An autonomous active learning framework is used to facilitate the exploration of designs.After six learning cycles and 2180 adsorption calculations across 15 intermediates,we develop a surrogate model that identifies four novel catalysts on the Pareto front of activity and selectivity.Our work demonstrates the effectiveness of leveraging a domain foundation model with an active learning framework and holds potential to significantly accelerate the discovery of high-performance CO_(2)RR catalysts.展开更多
Machine learning has emerged as a key approach in wildfire risk prediction research.However,in practical applications,the scarcity of data for specific regions often hindersmodel performance,with models trained on reg...Machine learning has emerged as a key approach in wildfire risk prediction research.However,in practical applications,the scarcity of data for specific regions often hindersmodel performance,with models trained on region-specific data struggling to generalize due to differences in data distributions.While traditional methods based on expert knowledge tend to generalize better across regions,they are limited in leveragingmulti-source data effectively,resulting in suboptimal predictive accuracy.This paper addresses this challenge by exploring how accumulated domain expertise in wildfire prediction can reduce model reliance on large volumes of high-quality data.An active learning algorithm is proposed based on XGBoost for wildfire risk assessment that autonomously identifies low-confidence predictions and seeks re-labeling through a human-in-the-loop or physics-based correction approach.This corrected data is reintegrated into the model,effectively preventing catastrophic forgetting.Experimental results demonstrate that the proposed human-in-the-loop approach significantly enhances labeling accuracy,predictive performance,and preserves the model's ability to generalize.These findings highlight the value of incorporating human expertise into machine learningmodels,offering a practical solution tomitigate data quality challenges and improvemodel reliability in wildfire risk prediction.展开更多
Dynamical systems often exhibit multiple attractors representing significantly different functioning conditions.A global map of attraction basins can offer valuable guidance for stabilizing or transitioning system sta...Dynamical systems often exhibit multiple attractors representing significantly different functioning conditions.A global map of attraction basins can offer valuable guidance for stabilizing or transitioning system states.Such a map can be constructed without prior system knowledge by identifying attractors across a sufficient number of points in the state space.However,determining the attractor for each initial state can be a laborious task.Here,we tackle the challenge of reconstructing attraction basins using as few initial points as possible.In each iteration of our approach,informative points are selected through random seeding and are driven along the current classification boundary,promoting the eventual selection of points that are both diverse and enlightening.The results across various experimental dynamical systems demonstrate that our approach requires fewer points than baseline methods while achieving comparable mapping accuracy.Additionally,the reconstructed map allows us to accurately estimate the minimum escape distance required to transition the system state to a target basin.展开更多
Human Activity Recognition(HAR)has become increasingly critical in civic surveillance,medical care monitoring,and institutional protection.Current deep learning-based approaches often suffer from excessive computation...Human Activity Recognition(HAR)has become increasingly critical in civic surveillance,medical care monitoring,and institutional protection.Current deep learning-based approaches often suffer from excessive computational complexity,limited generalizability under varying conditions,and compromised real-time performance.To counter these,this paper introduces an Active Learning-aided Heuristic Deep Spatio-Textural Ensemble Learning(ALH-DSEL)framework.The model initially identifies keyframes from the surveillance videos with a Multi-Constraint Active Learning(MCAL)approach,with features extracted from DenseNet121.The frames are then segmented employing an optimized Fuzzy C-Means clustering algorithm with Firefly to identify areas of interest.A deep ensemble feature extractor,comprising DenseNet121,EfficientNet-B7,MobileNet,and GLCM,extracts varied spatial and textural features.Fused characteristics are enhanced through PCA and Min-Max normalization and discriminated by a maximum voting ensemble of RF,AdaBoost,and XGBoost.The experimental results show that ALH-DSEL provides higher accuracy,precision,recall,and F1-score,validating its superiority for real-time HAR in surveillance scenarios.展开更多
To capture the nonlinear dynamics and gain evolution in chirped pulse amplification(CPA)systems,the split-step Fourier method and the fourth-order Runge–Kutta method are integrated to iteratively address the generali...To capture the nonlinear dynamics and gain evolution in chirped pulse amplification(CPA)systems,the split-step Fourier method and the fourth-order Runge–Kutta method are integrated to iteratively address the generalized nonlinear Schrödinger equation and the rate equations.However,this approach is burdened by substantial computational demands,resulting in significant time expenditures.In the context of intelligent laser optimization and inverse design,the necessity for numerous simulations further exacerbates this issue,highlighting the need for fast and accurate simulation methodologies.Here,we introduce an end-to-end model augmented with active learning(E2E-AL)with decent generalization through different dedicated embedding methods over various parameters.On an identical computational platform,the artificial intelligence–driven model is 2000 times faster than the conventional simulation method.Benefiting from the active learning strategy,the E2E-AL model achieves decent precision with only two-thirds of the training samples compared with the case without such a strategy.Furthermore,we demonstrate a multi-objective inverse design of the CPA systems enabled by the E2E-AL model.The E2E-AL framework manifests the potential of becoming a standard approach for the rapid and accurate modeling of ultrafast lasers and is readily extended to simulate other complex systems.展开更多
The Reliability-Based Design Optimization(RBDO)of complex engineering structures considering uncertainties has problems of being high-dimensional,highly nonlinear,and timeconsuming,which requires a significant amount ...The Reliability-Based Design Optimization(RBDO)of complex engineering structures considering uncertainties has problems of being high-dimensional,highly nonlinear,and timeconsuming,which requires a significant amount of sampling simulation computation.In this paper,a basis-adaptive Polynomial Chaos(PC)-Kriging surrogate model is proposed,in order to relieve the computational burden and enhance the predictive accuracy of a metamodel.The active learning basis-adaptive PC-Kriging model is combined with a quantile-based RBDO framework.Finally,five engineering cases have been implemented,including a benchmark RBDO problem,three high-dimensional explicit problems,and a high-dimensional implicit problem.Compared with Support Vector Regression(SVR),Kriging,and polynomial chaos expansion models,results show that the proposed basis-adaptive PC-Kriging model is more accurate and efficient for RBDO problems of complex engineering structures.展开更多
One-step direct production of methanol from methane and water(PMMW)under mild conditions is challenging in heterogeneous catalysis owing to the absence of highly effective catalysts.Herein,we designed a series of“Sin...One-step direct production of methanol from methane and water(PMMW)under mild conditions is challenging in heterogeneous catalysis owing to the absence of highly effective catalysts.Herein,we designed a series of“Single-Atom”-“Frustrated Lewis Pair”(SA-FLP)dual active sites for the direct PMMW via density functional theory(DFT)calculations combined with a machine learning(ML)approach.The results indicate that the nine designed SA-FLP catalysts are capable of efficiently activate CH4 and H_(2)O and facilitate the coupling of OH^(*)and CH_(3)^(*)into methanol.The DFT-based microkinetic simulation(MKM)results indicate that CH_(3)OH production on Co1-FLP and Pt1-FLP catalysts can reach the turnover frequencies(TOFs)of 1.01×10^(−3)s^(-1)and 8.80×10^(−4)s^(-1),respectively,which exceed the experimentally reported values by three orders of magnitude.ML results unveil that the gradient boosted regression model with 13 simple features could give satisfactory predictions for the TOFs of CH_(3)OH production with RMSE and R^(2)of 0.009 s^(-1)and 1.00,respectively.The ML-predicted MKM results indicate that four catalysts including V_(1-),Fe_(1-),Ti_(1-),and Mn_(1)-FLP exhibit higher TOFs of CH_(3)OH production than the value that the most relevant experiments reported,indicating that the four catalysts are also promising catalysts for the PMMW.This study not only develops a simple and efficient approach for design and screening SA-FLP catalysts but also provides mechanistic insights into the direct PMMW.展开更多
Objective:Deep learning(DL)has become the prevailing method in chest radiograph analysis,yet its performance heavily depends on large quantities of annotated images.To mitigate the cost,cold-start active learning(AL),...Objective:Deep learning(DL)has become the prevailing method in chest radiograph analysis,yet its performance heavily depends on large quantities of annotated images.To mitigate the cost,cold-start active learning(AL),comprising an initialization followed by subsequent learning,selects a small subset of informative data points for labeling.Recent advancements in pretrained models by supervised or self-supervised learning tailored to chest radiograph have shown broad applicability to diverse downstream tasks.However,their potential in cold-start AL remains unexplored.Methods:To validate the efficacy of domain-specific pretraining,we compared two foundation models:supervised TXRV and self-supervised REMEDIS with their general domain counterparts pretrained on ImageNet.Model performance was evaluated at both initialization and subsequent learning stages on two diagnostic tasks:psychiatric pneumonia and COVID-19.For initialization,we assessed their integration with three strategies:diversity,uncertainty,and hybrid sampling.For subsequent learning,we focused on uncertainty sampling powered by different pretrained models.We also conducted statistical tests to compare the foundation models with ImageNet counterparts,investigate the relationship between initialization and subsequent learning,examine the performance of one-shot initialization against the full AL process,and investigate the influence of class balance in initialization samples on initialization and subsequent learning.Results:First,domain-specific foundation models failed to outperform ImageNet counterparts in six out of eight experiments on informative sample selection.Both domain-specific and general pretrained models were unable to generate representations that could substitute for the original images as model inputs in seven of the eight scenarios.However,pretrained model-based initialization surpassed random sampling,the default approach in cold-start AL.Second,initialization performance was positively correlated with subsequent learning performance,highlighting the importance of initialization strategies.Third,one-shot initialization performed comparably to the full AL process,demonstrating the potential of reducing experts'repeated waiting during AL iterations.Last,a U-shaped correlation was observed between the class balance of initialization samples and model performance,suggesting that the class balance is more strongly associated with performance at middle budget levels than at low or high budgets.Conclusions:In this study,we highlighted the limitations of medical pretraining compared to general pretraining in the context of cold-start AL.We also identified promising outcomes related to cold-start AL,including initialization based on pretrained models,the positive influence of initialization on subsequent learning,the potential for one-shot initialization,and the influence of class balance on middle-budget AL.Researchers are encouraged to improve medical pretraining for versatile DL foundations and explore novel AL methods.展开更多
1.Colors of chemical reaction engineering models Kinetic models of chemical reactions are a crucial asset for understanding and optimizing chemical processes[1].These models are critical for reactor design,process opt...1.Colors of chemical reaction engineering models Kinetic models of chemical reactions are a crucial asset for understanding and optimizing chemical processes[1].These models are critical for reactor design,process optimization,catalyst design,scale-up,and process control,making them indispensable in the chemical industry.Kinetic models predict the change in temperature and concentration of the relevant species,given an actual concentration and temperature.Reaction predictions are made by integrating the kinetic model with a reactor model,which accounts for external constraints,such as flow,inlet concentration。展开更多
Support vector machines（SVMs） are a popular class of supervised learning algorithms, and are particularly applicable to large and high-dimensional classification problems. Like most machine learning methods for data...Support vector machines（SVMs） are a popular class of supervised learning algorithms, and are particularly applicable to large and high-dimensional classification problems. Like most machine learning methods for data classification and information retrieval, they require manually labeled data samples in the training stage. However, manual labeling is a time consuming and errorprone task. One possible solution to this issue is to exploit the large number of unlabeled samples that are easily accessible via the internet. This paper presents a novel active learning method for text categorization. The main objective of active learning is to reduce the labeling effort, without compromising the accuracy of classification, by intelligently selecting which samples should be labeled.The proposed method selects a batch of informative samples using the posterior probabilities provided by a set of multi-class SVM classifiers, and these samples are then manually labeled by an expert. Experimental results indicate that the proposed active learning method significantly reduces the labeling effort, while simultaneously enhancing the classification accuracy.展开更多
This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine (SVM) classifiers to learn an initial query concept. Then the proposed ac...This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine (SVM) classifiers to learn an initial query concept. Then the proposed active learning scheme employs similarity measure to check the current version space and selects images with maximum expected information gain to solicit user's label. Finally, the learned query is refined based on the user's further feedback. With the combination of SVM classifier and similarity measure, the proposed method can alleviate model bias existing in each of them. Our experiments on several query concepts show that the proposed method can learn the user's query concept quickly and effectively only with several iterations.展开更多
With the growing advancement of wireless communication technologies,WiFi-based human sensing has gained increasing attention as a non-intrusive and device-free solution.Among the available signal types,Channel State I...With the growing advancement of wireless communication technologies,WiFi-based human sensing has gained increasing attention as a non-intrusive and device-free solution.Among the available signal types,Channel State Information(CSI)offers fine-grained temporal,frequency,and spatial insights into multipath propagation,making it a crucial data source for human-centric sensing.Recently,the integration of deep learning has significantly improved the robustness and automation of feature extraction from CSI in complex environments.This paper provides a comprehensive review of deep learning-enhanced human sensing based on CSI.We first outline mainstream CSI acquisition tools and their hardware specifications,then provide a detailed discussion of preprocessing methods such as denoising,time–frequency transformation,data segmentation,and augmentation.Subsequently,we categorize deep learning approaches according to sensing tasks—namely detection,localization,and recognition—and highlight representative models across application scenarios.Finally,we examine key challenges including domain generalization,multi-user interference,and limited data availability,and we propose future research directions involving lightweight model deployment,multimodal data fusion,and semantic-level sensing.展开更多
Activity cliffs(ACs)are generally defined as pairs of similar compounds that only differ by a minor structural modification but exhibit a large difference in their binding affinity for a given target.ACs offer crucial...Activity cliffs(ACs)are generally defined as pairs of similar compounds that only differ by a minor structural modification but exhibit a large difference in their binding affinity for a given target.ACs offer crucial insights that aid medicinal chemists in optimizing molecular structures.Nonetheless,they also form a major source of prediction error in structure-activity relationship(SAR)models.To date,several studies have demonstrated that deep neural networks based on molecular images or graphs might need to be improved further in predicting the potency of ACs.In this paper,we integrated the triplet loss in face recognition with pre-training strategy to develop a prediction model ACtriplet,tailored for ACs.Through extensive comparison with multiple baseline models on 30 benchmark datasets,the results showed that ACtriplet was significantly better than those deep learning(DL)models without pretraining.In addition,we explored the effect of pre-training on data representation.Finally,the case study demonstrated that our model's interpretability module could explain the prediction results reasonably.In the dilemma that the amount of data could not be increased rapidly,this innovative framework would better make use of the existing data,which would propel the potential of DL in the early stage of drug discovery and optimization.展开更多
Education philosophy,which plays a major role in teacher beliefs,is crucial for teacher educators. Ever since the end of last century, there has been a trend in North America to promote, or restore active learning in ...Education philosophy,which plays a major role in teacher beliefs,is crucial for teacher educators. Ever since the end of last century, there has been a trend in North America to promote, or restore active learning in college/university classrooms. This paper, based on some explorations of the social constructivism, the six assumptions in Andragogy, the ARCS model, the active learning theory and some practical activities, proposes an integrated education philosophy about adult learning. It is assumed this will be able to provide some insight and guidance for college practitioners.展开更多
In active-learning classroom,the students play an active role in learning,and they are not only actively participating in classroom activities,but are interacting with the teacher and their peers.The teachers shoul tr...In active-learning classroom,the students play an active role in learning,and they are not only actively participating in classroom activities,but are interacting with the teacher and their peers.The teachers shoul try to avoid a noisy class with no real effects at all and they should help those who are poor at expressing in English.They should be encouraged to express themselves.To conclude,active-study method has brought us reward.展开更多
In the field of optoelectronics,certain types of data may be difficult to accurately annotate,such as high-resolution optoelectronic imaging or imaging in certain special spectral ranges.Weakly supervised learning can...In the field of optoelectronics,certain types of data may be difficult to accurately annotate,such as high-resolution optoelectronic imaging or imaging in certain special spectral ranges.Weakly supervised learning can provide a more reliable approach in these situations.Current popular approaches mainly adopt the classification-based class activation maps(CAM)as initial pseudo labels to solve the task.展开更多
Human Activity Recognition(HAR)represents a rapidly advancing research domain,propelled by continuous developments in sensor technologies and the Internet of Things(IoT).Deep learning has become the dominant paradigm ...Human Activity Recognition(HAR)represents a rapidly advancing research domain,propelled by continuous developments in sensor technologies and the Internet of Things(IoT).Deep learning has become the dominant paradigm in sensor-based HAR systems,offering significant advantages over traditional machine learning methods by eliminating manual feature extraction,enhancing recognition accuracy for complex activities,and enabling the exploitation of unlabeled data through generative models.This paper provides a comprehensive review of recent advancements and emerging trends in deep learning models developed for sensor-based human activity recognition(HAR)systems.We begin with an overview of fundamental HAR concepts in sensor-driven contexts,followed by a systematic categorization and summary of existing research.Our survey encompasses a wide range of deep learning approaches,including Multi-Layer Perceptrons(MLP),Convolutional Neural Networks(CNN),Recurrent Neural Networks(RNN),Long Short-Term Memory networks(LSTM),Gated Recurrent Units(GRU),Transformers,Deep Belief Networks(DBN),and hybrid architectures.A comparative evaluation of these models is provided,highlighting their performance,architectural complexity,and contributions to the field.Beyond Centralized deep learning models,we examine the role of Federated Learning(FL)in HAR,highlighting current applications and research directions.Finally,we discuss the growing importance of Explainable Artificial Intelligence(XAI)in sensor-based HAR,reviewing recent studies that integrate interpretability methods to enhance transparency and trustworthiness in deep learning-based HAR systems.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.12374253,12074053,12004064)J.G.thanks the Foreign talents project(G2022127004L),The authors also acknowledge computer support from the Shanghai Supercomputer Center,the DUT Supercomputing Center,and the Tianhe supercomputer of Tianjin Center.
文摘Surface-supported clusters forming by aggregation of excessive adatoms could be the main defects of 2D materials after chemical vapor deposition.They will significantly impact the electronic/magnetic properties.Moreover,surface supported atoms are also widely explored for high active and selecting catalysts.Severe deformation,even dipping into the surface,of these clusters can be expected because of the very active edge of clusters and strong interaction between supported clusters and surfaces.However,most models of these clusters are supposed to simply float on the top of the surface because ab initio simulations cannot afford the complex reconstructions.Here,we develop an accurate graph neural network machine learning potential(MLP)from ab initio data by active learning architecture through fine-tuning pre-trained models,and then employ the MLP into Monte Carlo to explore the structural evolutions of Mo and S clusters(1-8 atoms)on perfect and various defective MoS2 monolayers.Interestingly,Mo clusters can always sink and embed themselves into MoS2 layers.In contrast,S clusters float on perfect surfaces.On the defective surface,a few S atoms will fill the vacancy and rest S clusters float on the top.Such significant structural reconstructions should be carefully taken into account.
基金supported by the Natural Science Foundation of China[grant numbers 52302093]Natural Science Foundation of Jiangxi Province[grant numbers 20224BAB204021].
文摘In materials science,a significant correlation often exists between material input parameters and their corresponding performance attributes.Nevertheless,the inherent challenges associated with small data obscure these statistical correlations,impeding machine learning models from effectively capturing the underlying patterns,thereby hampering efficient optimization of material properties.This work presents a novel active learning framework that integrates generative adversarial networks(GAN)with a directionally constrained expected absolute improvement(EAI)acquisition function to accelerate the discovery of ultra-high temperature ceramics(UHTCs)using small data.The framework employs GAN for data augmentation,symbolic regression for feature weight derivation,and a self-developed EAI function that incorporates input feature importance weighting to quantify bidirectional deviations from zero ablation rate.Through only two iterations,this framework successfully identified the optimal composition of HfB_(2)-3.52SiC-5.23TaSi_(2),which exhibits robust near-zero ablation rates under plasma ablation at 2500℃ for 200 s,demonstrating superior sampling efficiency compared to conventional active learning approaches.Microstructural analysis reveals that the exceptional performance stems from the formation of a highly viscous HfO_(2)-SiO_(2)-Ta_(2)O_(5)-HfSiO_(4)-Hf_(3)(BO_(3))_(4) oxide layer,which provides effective oxygen barrier protection.This work demonstrates an efficient and universal approach for rapid materials discovery using small data.
基金supported by the National Key Research and Development Program of China(2022ZD0117501)the Scientific Research Innovation Capability Support Project for Young Faculty(ZYGXQNJSKYCXNLZCXM-E7)the Tsinghua University Initiative Scientific Research Program and the Carbon Neutrality and Energy System Transformation(CNEST)Program led by Tsinghua University.
文摘Industrial decarbonization is critical for achieving net-zero goals.The carbon dioxide electrochemical reduction reaction(CO_(2)RR)is a promising approach for converting CO_(2)into high-value chemicals,offering the potential for decarbonizing industrial processes toward a sustainable,carbon-neutral future.However,developing CO_(2)RR catalysts with high selectivity and activity remains a challenge due to the complexity of finding such catalysts and the inefficiency of traditional computational or experimental approaches.Here,we present a methodology integrating density functional theory(DFT)calculations,deep learning models,and an active learning strategy to rapidly screen high-performance catalysts.The proposed methodology is then demonstrated on graphene-based single-atom catalysts for selective CO_(2)electroreduction to methanol.First,we conduct systematic binding energy calculations for 3045 single-atom catalysts to identify thermodynamically stable catalysts as the design space.We then use a graph neural network,fine-tuned with a specialized adsorption energy database,to predict the relative activity and selectivity of the candidate catalysts.An autonomous active learning framework is used to facilitate the exploration of designs.After six learning cycles and 2180 adsorption calculations across 15 intermediates,we develop a surrogate model that identifies four novel catalysts on the Pareto front of activity and selectivity.Our work demonstrates the effectiveness of leveraging a domain foundation model with an active learning framework and holds potential to significantly accelerate the discovery of high-performance CO_(2)RR catalysts.
基金funded by the Natural Science Foundation of Guangxi Province(Grant AB24010157)Research Project of the Sichuan Forestry and Grassland Bureau(Grants G202206012 and G202206012-2)+1 种基金National Natural Science Foundation of China(Grants 32471878,62373081,U2330206,U2230206 and 62173068)Sichuan Science and Technology Program(Grants 2024NSFSC1483,2024ZYD0156,2023NSFC1962 and DQ202412).
文摘Machine learning has emerged as a key approach in wildfire risk prediction research.However,in practical applications,the scarcity of data for specific regions often hindersmodel performance,with models trained on region-specific data struggling to generalize due to differences in data distributions.While traditional methods based on expert knowledge tend to generalize better across regions,they are limited in leveragingmulti-source data effectively,resulting in suboptimal predictive accuracy.This paper addresses this challenge by exploring how accumulated domain expertise in wildfire prediction can reduce model reliance on large volumes of high-quality data.An active learning algorithm is proposed based on XGBoost for wildfire risk assessment that autonomously identifies low-confidence predictions and seeks re-labeling through a human-in-the-loop or physics-based correction approach.This corrected data is reintegrated into the model,effectively preventing catastrophic forgetting.Experimental results demonstrate that the proposed human-in-the-loop approach significantly enhances labeling accuracy,predictive performance,and preserves the model's ability to generalize.These findings highlight the value of incorporating human expertise into machine learningmodels,offering a practical solution tomitigate data quality challenges and improvemodel reliability in wildfire risk prediction.
基金supported by the National Natural Science Foundation of China(Grant Nos.T2225022,12350710786,62088101,and 12161141016)Shuguang Program of Shanghai Education Development Foundation(Grant No.22SG21)Shanghai Municipal Education Commission,and the Fundamental Research Funds for the Central Universities。
文摘Dynamical systems often exhibit multiple attractors representing significantly different functioning conditions.A global map of attraction basins can offer valuable guidance for stabilizing or transitioning system states.Such a map can be constructed without prior system knowledge by identifying attractors across a sufficient number of points in the state space.However,determining the attractor for each initial state can be a laborious task.Here,we tackle the challenge of reconstructing attraction basins using as few initial points as possible.In each iteration of our approach,informative points are selected through random seeding and are driven along the current classification boundary,promoting the eventual selection of points that are both diverse and enlightening.The results across various experimental dynamical systems demonstrate that our approach requires fewer points than baseline methods while achieving comparable mapping accuracy.Additionally,the reconstructed map allows us to accurately estimate the minimum escape distance required to transition the system state to a target basin.
文摘Human Activity Recognition(HAR)has become increasingly critical in civic surveillance,medical care monitoring,and institutional protection.Current deep learning-based approaches often suffer from excessive computational complexity,limited generalizability under varying conditions,and compromised real-time performance.To counter these,this paper introduces an Active Learning-aided Heuristic Deep Spatio-Textural Ensemble Learning(ALH-DSEL)framework.The model initially identifies keyframes from the surveillance videos with a Multi-Constraint Active Learning(MCAL)approach,with features extracted from DenseNet121.The frames are then segmented employing an optimized Fuzzy C-Means clustering algorithm with Firefly to identify areas of interest.A deep ensemble feature extractor,comprising DenseNet121,EfficientNet-B7,MobileNet,and GLCM,extracts varied spatial and textural features.Fused characteristics are enhanced through PCA and Min-Max normalization and discriminated by a maximum voting ensemble of RF,AdaBoost,and XGBoost.The experimental results show that ALH-DSEL provides higher accuracy,precision,recall,and F1-score,validating its superiority for real-time HAR in surveillance scenarios.
基金supported by the National Natural Science Foundation of China(Grant Nos.62227821,62025503,and 62205199).
文摘To capture the nonlinear dynamics and gain evolution in chirped pulse amplification(CPA)systems,the split-step Fourier method and the fourth-order Runge–Kutta method are integrated to iteratively address the generalized nonlinear Schrödinger equation and the rate equations.However,this approach is burdened by substantial computational demands,resulting in significant time expenditures.In the context of intelligent laser optimization and inverse design,the necessity for numerous simulations further exacerbates this issue,highlighting the need for fast and accurate simulation methodologies.Here,we introduce an end-to-end model augmented with active learning(E2E-AL)with decent generalization through different dedicated embedding methods over various parameters.On an identical computational platform,the artificial intelligence–driven model is 2000 times faster than the conventional simulation method.Benefiting from the active learning strategy,the E2E-AL model achieves decent precision with only two-thirds of the training samples compared with the case without such a strategy.Furthermore,we demonstrate a multi-objective inverse design of the CPA systems enabled by the E2E-AL model.The E2E-AL framework manifests the potential of becoming a standard approach for the rapid and accurate modeling of ultrafast lasers and is readily extended to simulate other complex systems.
基金supported by the National Key R&D Program of China(No.2021YFB1715000)the National Natural Science Foundation of China(No.52375073)。
文摘The Reliability-Based Design Optimization(RBDO)of complex engineering structures considering uncertainties has problems of being high-dimensional,highly nonlinear,and timeconsuming,which requires a significant amount of sampling simulation computation.In this paper,a basis-adaptive Polynomial Chaos(PC)-Kriging surrogate model is proposed,in order to relieve the computational burden and enhance the predictive accuracy of a metamodel.The active learning basis-adaptive PC-Kriging model is combined with a quantile-based RBDO framework.Finally,five engineering cases have been implemented,including a benchmark RBDO problem,three high-dimensional explicit problems,and a high-dimensional implicit problem.Compared with Support Vector Regression(SVR),Kriging,and polynomial chaos expansion models,results show that the proposed basis-adaptive PC-Kriging model is more accurate and efficient for RBDO problems of complex engineering structures.
文摘One-step direct production of methanol from methane and water(PMMW)under mild conditions is challenging in heterogeneous catalysis owing to the absence of highly effective catalysts.Herein,we designed a series of“Single-Atom”-“Frustrated Lewis Pair”(SA-FLP)dual active sites for the direct PMMW via density functional theory(DFT)calculations combined with a machine learning(ML)approach.The results indicate that the nine designed SA-FLP catalysts are capable of efficiently activate CH4 and H_(2)O and facilitate the coupling of OH^(*)and CH_(3)^(*)into methanol.The DFT-based microkinetic simulation(MKM)results indicate that CH_(3)OH production on Co1-FLP and Pt1-FLP catalysts can reach the turnover frequencies(TOFs)of 1.01×10^(−3)s^(-1)and 8.80×10^(−4)s^(-1),respectively,which exceed the experimentally reported values by three orders of magnitude.ML results unveil that the gradient boosted regression model with 13 simple features could give satisfactory predictions for the TOFs of CH_(3)OH production with RMSE and R^(2)of 0.009 s^(-1)and 1.00,respectively.The ML-predicted MKM results indicate that four catalysts including V_(1-),Fe_(1-),Ti_(1-),and Mn_(1)-FLP exhibit higher TOFs of CH_(3)OH production than the value that the most relevant experiments reported,indicating that the four catalysts are also promising catalysts for the PMMW.This study not only develops a simple and efficient approach for design and screening SA-FLP catalysts but also provides mechanistic insights into the direct PMMW.
文摘Objective:Deep learning(DL)has become the prevailing method in chest radiograph analysis,yet its performance heavily depends on large quantities of annotated images.To mitigate the cost,cold-start active learning(AL),comprising an initialization followed by subsequent learning,selects a small subset of informative data points for labeling.Recent advancements in pretrained models by supervised or self-supervised learning tailored to chest radiograph have shown broad applicability to diverse downstream tasks.However,their potential in cold-start AL remains unexplored.Methods:To validate the efficacy of domain-specific pretraining,we compared two foundation models:supervised TXRV and self-supervised REMEDIS with their general domain counterparts pretrained on ImageNet.Model performance was evaluated at both initialization and subsequent learning stages on two diagnostic tasks:psychiatric pneumonia and COVID-19.For initialization,we assessed their integration with three strategies:diversity,uncertainty,and hybrid sampling.For subsequent learning,we focused on uncertainty sampling powered by different pretrained models.We also conducted statistical tests to compare the foundation models with ImageNet counterparts,investigate the relationship between initialization and subsequent learning,examine the performance of one-shot initialization against the full AL process,and investigate the influence of class balance in initialization samples on initialization and subsequent learning.Results:First,domain-specific foundation models failed to outperform ImageNet counterparts in six out of eight experiments on informative sample selection.Both domain-specific and general pretrained models were unable to generate representations that could substitute for the original images as model inputs in seven of the eight scenarios.However,pretrained model-based initialization surpassed random sampling,the default approach in cold-start AL.Second,initialization performance was positively correlated with subsequent learning performance,highlighting the importance of initialization strategies.Third,one-shot initialization performed comparably to the full AL process,demonstrating the potential of reducing experts'repeated waiting during AL iterations.Last,a U-shaped correlation was observed between the class balance of initialization samples and model performance,suggesting that the class balance is more strongly associated with performance at middle budget levels than at low or high budgets.Conclusions:In this study,we highlighted the limitations of medical pretraining compared to general pretraining in the context of cold-start AL.We also identified promising outcomes related to cold-start AL,including initialization based on pretrained models,the positive influence of initialization on subsequent learning,the potential for one-shot initialization,and the influence of class balance on middle-budget AL.Researchers are encouraged to improve medical pretraining for versatile DL foundations and explore novel AL methods.
基金Yannick Ureel and Maarten Dobbelaere acknowledge financial support from the Fund for Scientific Research Flanders(FWO Flanders)respectively through doctoral fellowship grants(1185822N and 1S45522N)The authors acknowledge funding from the European Research Council under the European Union’s Horizon 2020 research and innovation programme/ERC(818607).
文摘1.Colors of chemical reaction engineering models Kinetic models of chemical reactions are a crucial asset for understanding and optimizing chemical processes[1].These models are critical for reactor design,process optimization,catalyst design,scale-up,and process control,making them indispensable in the chemical industry.Kinetic models predict the change in temperature and concentration of the relevant species,given an actual concentration and temperature.Reaction predictions are made by integrating the kinetic model with a reactor model,which accounts for external constraints,such as flow,inlet concentration。
文摘Support vector machines（SVMs） are a popular class of supervised learning algorithms, and are particularly applicable to large and high-dimensional classification problems. Like most machine learning methods for data classification and information retrieval, they require manually labeled data samples in the training stage. However, manual labeling is a time consuming and errorprone task. One possible solution to this issue is to exploit the large number of unlabeled samples that are easily accessible via the internet. This paper presents a novel active learning method for text categorization. The main objective of active learning is to reduce the labeling effort, without compromising the accuracy of classification, by intelligently selecting which samples should be labeled.The proposed method selects a batch of informative samples using the posterior probabilities provided by a set of multi-class SVM classifiers, and these samples are then manually labeled by an expert. Experimental results indicate that the proposed active learning method significantly reduces the labeling effort, while simultaneously enhancing the classification accuracy.
文摘This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine (SVM) classifiers to learn an initial query concept. Then the proposed active learning scheme employs similarity measure to check the current version space and selects images with maximum expected information gain to solicit user's label. Finally, the learned query is refined based on the user's further feedback. With the combination of SVM classifier and similarity measure, the proposed method can alleviate model bias existing in each of them. Our experiments on several query concepts show that the proposed method can learn the user's query concept quickly and effectively only with several iterations.
基金supported by National Natural Science Foundation of China(NSFC)under grant U23A20310.
文摘With the growing advancement of wireless communication technologies,WiFi-based human sensing has gained increasing attention as a non-intrusive and device-free solution.Among the available signal types,Channel State Information(CSI)offers fine-grained temporal,frequency,and spatial insights into multipath propagation,making it a crucial data source for human-centric sensing.Recently,the integration of deep learning has significantly improved the robustness and automation of feature extraction from CSI in complex environments.This paper provides a comprehensive review of deep learning-enhanced human sensing based on CSI.We first outline mainstream CSI acquisition tools and their hardware specifications,then provide a detailed discussion of preprocessing methods such as denoising,time–frequency transformation,data segmentation,and augmentation.Subsequently,we categorize deep learning approaches according to sensing tasks—namely detection,localization,and recognition—and highlight representative models across application scenarios.Finally,we examine key challenges including domain generalization,multi-user interference,and limited data availability,and we propose future research directions involving lightweight model deployment,multimodal data fusion,and semantic-level sensing.
基金supported by the National Natural Science Foundation of China(Grant Nos.:U23A20530,82273858,and 82173746)the National Key Research and Development Programof China(Grant No.:2023YFF1204904)Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism(Shanghai Municipal Education Commission,China).
文摘Activity cliffs(ACs)are generally defined as pairs of similar compounds that only differ by a minor structural modification but exhibit a large difference in their binding affinity for a given target.ACs offer crucial insights that aid medicinal chemists in optimizing molecular structures.Nonetheless,they also form a major source of prediction error in structure-activity relationship(SAR)models.To date,several studies have demonstrated that deep neural networks based on molecular images or graphs might need to be improved further in predicting the potency of ACs.In this paper,we integrated the triplet loss in face recognition with pre-training strategy to develop a prediction model ACtriplet,tailored for ACs.Through extensive comparison with multiple baseline models on 30 benchmark datasets,the results showed that ACtriplet was significantly better than those deep learning(DL)models without pretraining.In addition,we explored the effect of pre-training on data representation.Finally,the case study demonstrated that our model's interpretability module could explain the prediction results reasonably.In the dilemma that the amount of data could not be increased rapidly,this innovative framework would better make use of the existing data,which would propel the potential of DL in the early stage of drug discovery and optimization.
文摘Education philosophy,which plays a major role in teacher beliefs,is crucial for teacher educators. Ever since the end of last century, there has been a trend in North America to promote, or restore active learning in college/university classrooms. This paper, based on some explorations of the social constructivism, the six assumptions in Andragogy, the ARCS model, the active learning theory and some practical activities, proposes an integrated education philosophy about adult learning. It is assumed this will be able to provide some insight and guidance for college practitioners.
文摘In active-learning classroom,the students play an active role in learning,and they are not only actively participating in classroom activities,but are interacting with the teacher and their peers.The teachers shoul try to avoid a noisy class with no real effects at all and they should help those who are poor at expressing in English.They should be encouraged to express themselves.To conclude,active-study method has brought us reward.
文摘In the field of optoelectronics,certain types of data may be difficult to accurately annotate,such as high-resolution optoelectronic imaging or imaging in certain special spectral ranges.Weakly supervised learning can provide a more reliable approach in these situations.Current popular approaches mainly adopt the classification-based class activation maps(CAM)as initial pseudo labels to solve the task.
文摘Human Activity Recognition(HAR)represents a rapidly advancing research domain,propelled by continuous developments in sensor technologies and the Internet of Things(IoT).Deep learning has become the dominant paradigm in sensor-based HAR systems,offering significant advantages over traditional machine learning methods by eliminating manual feature extraction,enhancing recognition accuracy for complex activities,and enabling the exploitation of unlabeled data through generative models.This paper provides a comprehensive review of recent advancements and emerging trends in deep learning models developed for sensor-based human activity recognition(HAR)systems.We begin with an overview of fundamental HAR concepts in sensor-driven contexts,followed by a systematic categorization and summary of existing research.Our survey encompasses a wide range of deep learning approaches,including Multi-Layer Perceptrons(MLP),Convolutional Neural Networks(CNN),Recurrent Neural Networks(RNN),Long Short-Term Memory networks(LSTM),Gated Recurrent Units(GRU),Transformers,Deep Belief Networks(DBN),and hybrid architectures.A comparative evaluation of these models is provided,highlighting their performance,architectural complexity,and contributions to the field.Beyond Centralized deep learning models,we examine the role of Federated Learning(FL)in HAR,highlighting current applications and research directions.Finally,we discuss the growing importance of Explainable Artificial Intelligence(XAI)in sensor-based HAR,reviewing recent studies that integrate interpretability methods to enhance transparency and trustworthiness in deep learning-based HAR systems.
