Heart monitoring improves life quality.Electrocardiograms(ECGs or EKGs)detect heart irregularities.Machine learning algorithms can create a few ECG diagnosis processing methods.The first method uses raw ECG and time-s...Heart monitoring improves life quality.Electrocardiograms(ECGs or EKGs)detect heart irregularities.Machine learning algorithms can create a few ECG diagnosis processing methods.The first method uses raw ECG and time-series data.The second method classifies the ECG by patient experience.The third technique translates ECG impulses into Q waves,R waves and S waves(QRS)features using richer information.Because ECG signals vary naturally between humans and activities,we will combine the three feature selection methods to improve classification accuracy and diagnosis.Classifications using all three approaches have not been examined till now.Several researchers found that Machine Learning(ML)techniques can improve ECG classification.This study will compare popular machine learning techniques to evaluate ECG features.Four algorithms—Support Vector Machine(SVM),Decision Tree,Naive Bayes,and Neural Network—compare categorization results.SVM plus prior knowledge has the highest accuracy(99%)of the four ML methods.QRS characteristics failed to identify signals without chaos theory.With 99.8%classification accuracy,the Decision Tree technique outperformed all previous experiments.展开更多
As a representative emerging machine learning technique, federated learning(FL) has gained considerable popularity for its special feature of “making data available but not visible”. However, potential problems rema...As a representative emerging machine learning technique, federated learning(FL) has gained considerable popularity for its special feature of “making data available but not visible”. However, potential problems remain, including privacy breaches, imbalances in payment, and inequitable distribution.These shortcomings let devices reluctantly contribute relevant data to, or even refuse to participate in FL. Therefore, in the application of FL, an important but also challenging issue is to motivate as many participants as possible to provide high-quality data to FL. In this paper, we propose an incentive mechanism for FL based on the continuous zero-determinant(CZD) strategies from the perspective of game theory. We first model the interaction between the server and the devices during the FL process as a continuous iterative game. We then apply the CZD strategies for two players and then multiple players to optimize the social welfare of FL, for which we prove that the server can keep social welfare at a high and stable level. Subsequently, we design an incentive mechanism based on the CZD strategies to attract devices to contribute all of their high-accuracy data to FL.Finally, we perform simulations to demonstrate that our proposed CZD-based incentive mechanism can indeed generate high and stable social welfare in FL.展开更多
Magnesium(Mg)alloys have shown great prospects as both structural and biomedical materials,while poor corrosion resistance limits their further application.In this work,to avoid the time-consuming and laborious experi...Magnesium(Mg)alloys have shown great prospects as both structural and biomedical materials,while poor corrosion resistance limits their further application.In this work,to avoid the time-consuming and laborious experiment trial,a high-throughput computational strategy based on first-principles calculations is designed for screening corrosion-resistant binary Mg alloy with intermetallics,from both the thermodynamic and kinetic perspectives.The stable binary Mg intermetallics with low equilibrium potential difference with respect to the Mg matrix are firstly identified.Then,the hydrogen adsorption energies on the surfaces of these Mg intermetallics are calculated,and the corrosion exchange current density is further calculated by a hydrogen evolution reaction(HER)kinetic model.Several intermetallics,e.g.Y_(3)Mg,Y_(2)Mg and La_(5)Mg,are identified to be promising intermetallics which might effectively hinder the cathodic HER.Furthermore,machine learning(ML)models are developed to predict Mg intermetallics with proper hydrogen adsorption energy employing work function(W_(f))and weighted first ionization energy(WFIE).The generalization of the ML models is tested on five new binary Mg intermetallics with the average root mean square error(RMSE)of 0.11 eV.This study not only predicts some promising binary Mg intermetallics which may suppress the galvanic corrosion,but also provides a high-throughput screening strategy and ML models for the design of corrosion-resistant alloy,which can be extended to ternary Mg alloys or other alloy systems.展开更多
The high throughput prediction of the thermodynamic phase behavior of active pharmaceutical ingredients(APIs)with pharmaceutically relevant excipients remains a major scientific challenge in the screening of pharmaceu...The high throughput prediction of the thermodynamic phase behavior of active pharmaceutical ingredients(APIs)with pharmaceutically relevant excipients remains a major scientific challenge in the screening of pharmaceutical formulations.In this work,a developed machine-learning model efficiently predicts the solubility of APIs in polymers by learning the phase equilibrium principle and using a few molecular descriptors.Under the few-shot learning framework,thermodynamic theory(perturbed-chain statistical associating fluid theory)was used for data augmentation,and computational chemistry was applied for molecular descriptors'screening.The results showed that the developed machine-learning model can predict the API-polymer phase diagram accurately,broaden the solubility data of APIs in polymers,and reproduce the relationship between API solubility and the interaction mechanisms between API and polymer successfully,which provided efficient guidance for the development of pharmaceutical formulations.展开更多
A comprehensive and precise analysis of shale gas production performance is crucial for evaluating resource potential,designing a field development plan,and making investment decisions.However,quantitative analysis ca...A comprehensive and precise analysis of shale gas production performance is crucial for evaluating resource potential,designing a field development plan,and making investment decisions.However,quantitative analysis can be challenging because production performance is dominated by the complex interaction among a series of geological and engineering factors.In fact,each factor can be viewed as a player who makes cooperative contributions to the production payoff within the constraints of physical laws and models.Inspired by the idea,we propose a hybrid data-driven analysis framework in this study,where the contributions of dominant factors are quantitatively evaluated,the productions are precisely forecasted,and the development optimization suggestions are comprehensively generated.More specifically,game theory and machine learning models are coupled to determine the dominating geological and engineering factors.The Shapley value with definite physical meaning is employed to quantitatively measure the effects of individual factors.A multi-model-fused stacked model is trained for production forecast,which provides the basis for derivative-free optimization algorithms to optimize the development plan.The complete workflow is validated with actual production data collected from the Fuling shale gas field,Sichuan Basin,China.The validation results show that the proposed procedure can draw rigorous conclusions with quantified evidence and thereby provide specific and reliable suggestions for development plan optimization.Comparing with traditional and experience-based approaches,the hybrid data-driven procedure is advanced in terms of both efficiency and accuracy.展开更多
Deep multi-modal learning,a rapidly growing field with a wide range of practical applications,aims to effectively utilize and integrate information from multiple sources,known as modalities.Despite its impressive empi...Deep multi-modal learning,a rapidly growing field with a wide range of practical applications,aims to effectively utilize and integrate information from multiple sources,known as modalities.Despite its impressive empirical performance,the theoretical foundations of deep multi-modal learning have yet to be fully explored.In this paper,we will undertake a comprehensive survey of recent developments in multi-modal learning theories,focusing on the fundamental properties that govern this field.Our goal is to provide a thorough collection of current theoretical tools for analyzing multi-modal learning,to clarify their implications for practitioners,and to suggest future directions for the establishment of a solid theoretical foundation for deep multi-modal learning.展开更多
In real-time strategy(RTS)games,the ability of recognizing other players’goals is important for creating artifical intelligence(AI)players.However,most current goal recognition methods do not take the player’s decep...In real-time strategy(RTS)games,the ability of recognizing other players’goals is important for creating artifical intelligence(AI)players.However,most current goal recognition methods do not take the player’s deceptive behavior into account which often occurs in RTS game scenarios,resulting in poor recognition results.In order to solve this problem,this paper proposes goal recognition for deceptive agent,which is an extended goal recognition method applying the deductive reason method(from general to special)to model the deceptive agent’s behavioral strategy.First of all,the general deceptive behavior model is proposed to abstract features of deception,and then these features are applied to construct a behavior strategy that best matches the deceiver’s historical behavior data by the inverse reinforcement learning(IRL)method.Final,to interfere with the deceptive behavior implementation,we construct a game model to describe the confrontation scenario and the most effective interference measures.展开更多
Attacks on the cyber space is getting exponential in recent times.Illegal penetrations and breaches are real threats to the individuals and organizations.Conventional security systems are good enough to detect the kno...Attacks on the cyber space is getting exponential in recent times.Illegal penetrations and breaches are real threats to the individuals and organizations.Conventional security systems are good enough to detect the known threats but when it comes to Advanced Persistent Threats(APTs)they fails.These APTs are targeted,more sophisticated and very persistent and incorporates lot of evasive techniques to bypass the existing defenses.Hence,there is a need for an effective defense system that can achieve a complete reliance of security.To address the above-mentioned issues,this paper proposes a novel honeypot system that tracks the anonymous behavior of the APT threats.The key idea of honeypot leverages the concepts of graph theory to detect such targeted attacks.The proposed honey-pot is self-realizing,strategic assisted which withholds the APTs actionable tech-niques and observes the behavior for analysis and modelling.The proposed graph theory based self learning honeypot using the resultsγ(C(n,1)),γc(C(n,1)),γsc(C(n,1))outperforms traditional techniques by detecting APTs behavioral with detection rate of 96%.展开更多
This study focuses on the effectiveness of the project-based language learning(PBLL) in a college Secretarial Oral English(SOE) Module. Student reflections of the language project work have been analyzed through Activ...This study focuses on the effectiveness of the project-based language learning(PBLL) in a college Secretarial Oral English(SOE) Module. Student reflections of the language project work have been analyzed through Activity Theory. Moreover,Data has been collected and categorized based on the components of complex human activity: the subject, object, tools(signs,symbols, and language), the community in which the activity take place, division of labor, and rules. The findings theoretically support the outcome of project-based language learning which align with the object of the activity.展开更多
The unceasing revolution of the global economy and culture boosts the revolutionary step of the educational circle.Combining the creed of The Guide of College English Teaching in 2016 with the results of investigation...The unceasing revolution of the global economy and culture boosts the revolutionary step of the educational circle.Combining the creed of The Guide of College English Teaching in 2016 with the results of investigation and survey in colleges, a research group in the Institute of Foreign Languages of Hankou University comes up with a revolutionary trial scheme on College English teaching conducted by discovery learning theory, as well as a research method of action research, which is in hope of mending the problems and shortcomings of current College English teaching.展开更多
Since traditional English teaching method, which merely focuses on language teaching but ignores communicative competence, severely impedes the development of students' oral ability. It is high time that English t...Since traditional English teaching method, which merely focuses on language teaching but ignores communicative competence, severely impedes the development of students' oral ability. It is high time that English teachers took measures to find a workable and valuable teaching method which can improve students' speaking proficiency effectively. Learning community theory provides a broad space for this, for it regards learning as a process which takes place in a community where the learners are sharing their experience towards knowledge building in an interactive and cooperative way.展开更多
This paper tries to summarize some main schools 0f teaching methodologies abroad and some main learning theories abroad. From this paper, we can know the main learning theories, the basic theories of them and the lead...This paper tries to summarize some main schools 0f teaching methodologies abroad and some main learning theories abroad. From this paper, we can know the main learning theories, the basic theories of them and the leading figures. It can help us understand the characteristics of each school of the teaching methodologies and learning theories.展开更多
Lithium has been paid great attention in recent years thanks to its significant appli-cations for battery and lightweight alloy.Developing a potential model with high ac-curacy and efficiency is impor-tant for theoret...Lithium has been paid great attention in recent years thanks to its significant appli-cations for battery and lightweight alloy.Developing a potential model with high ac-curacy and efficiency is impor-tant for theoretical simulation of lithium materials.Here,we build a deep learning potential(DP)for elemental lithium based on a concurrent-learning scheme and DP representation of the density-functional theory(DFT)potential energy surface(PES),the DP model enables material simulations with close-to DFT accuracy but at much lower computational cost.The simulations show that basic parameters,equation of states,elasticity,defects and surface are consistent with the first principles results.More notably,the liquid radial distribution func-tion based on our DP model is found to match well with experiment data.Our results demon-strate that the developed DP model can be used for the simulation of lithium materials.展开更多
Social engineering attacks are considered one of the most hazardous cyberattacks in cybersecurity,as human vulnerabilities are often the weakest link in the entire network.Such vulnerabilities are becoming increasingl...Social engineering attacks are considered one of the most hazardous cyberattacks in cybersecurity,as human vulnerabilities are often the weakest link in the entire network.Such vulnerabilities are becoming increasingly susceptible to network security risks.Addressing the social engineering attack defense problem has been the focus of many studies.However,two main challenges hinder its successful resolution.Firstly,the vulnerabilities in social engineering attacks are unique due to multistage attacks,leading to incorrect social engineering defense strategies.Secondly,social engineering attacks are real-time,and the defense strategy algorithms based on gaming or reinforcement learning are too complex to make rapid decisions.This paper proposes a multiattribute quantitative incentive method based on human vulnerability and an improved Q-learning(IQL)reinforcement learning method on human vulnerability attributes.The proposed algorithm aims to address the two main challenges in social engineering attack defense by using a multiattribute incentive method based on human vulnerability to determine the optimal defense strategy.Furthermore,the IQL reinforcement learning method facilitates rapid decision-making during real-time attacks.The experimental results demonstrate that the proposed algorithm outperforms the traditional Qlearning(QL)and deep Q-network(DQN)approaches in terms of time efficiency,taking 9.1%and 19.4%less time,respectively.Moreover,the proposed algorithm effectively addresses the non-uniformity of vulnerabilities in social engineering attacks and provides a reliable defense strategy based on human vulnerability attributes.This study contributes to advancing social engineering attack defense by introducing an effective and efficient method for addressing the vulnerabilities of human factors in the cybersecurity domain.展开更多
Second language acquisition can not be understood without addressing the interaction between language and cognition. Cognitive theory can extend to describe learning strategies as complex cognitive skills. Theoretical...Second language acquisition can not be understood without addressing the interaction between language and cognition. Cognitive theory can extend to describe learning strategies as complex cognitive skills. Theoretical developments in Anderson’s production systems cover a broader range of behavior than other theories, including comprehension and production of oral and written texts as well as comprehension, problem solving, and verbal learning.Thus Anderson’s cognitive theory can be served as a rationale for learning strategy studies in second language acquisition.展开更多
By analyzing the English learning logs of 12 students in a provincial university in south-west China after they had been exempted from taking college English courses,this study investigated college students’autonomou...By analyzing the English learning logs of 12 students in a provincial university in south-west China after they had been exempted from taking college English courses,this study investigated college students’autonomous EFL(English as a foreign language)learning after course exemption,including the use of mediational means in EFL learning,EFL learning hours,and other factors affecting EFL learning,in the hope of giving new perspectives on college ELF curriculum design,teaching,and education management.展开更多
This systematic review is focused on the importance of the Bandura Social Cognitive Theory and its theoretical components such as self-efficacy in workplace of today.The themes have been described through the course o...This systematic review is focused on the importance of the Bandura Social Cognitive Theory and its theoretical components such as self-efficacy in workplace of today.The themes have been described through the course of the decades.These theories have been utilized heavily in research and real-life case studies,have been further developed by Bandura and other researchers,and have been implemented in organizational psychology.During the past 10 years they have helped in reshaping Human Resources Development.Major latest contributions and applications are discussed touching even the recent outbreak of the pandemic.The influence of the theory is immense and the importance of self-efficacy in the workplace has been addressed and proven by research.展开更多
Based on a new theoretical perspective, this paper attempts to unify the seemingly incompatible learning theories of Connectivism and Constructivism into a scientific theoretical framework. The Connectivism-Constructi...Based on a new theoretical perspective, this paper attempts to unify the seemingly incompatible learning theories of Connectivism and Constructivism into a scientific theoretical framework. The Connectivism-Constructivism learning theory is not a simple superposition of the two theories. Instead, it absorbs the essence of the learning theory of Constructivism, Connectivism and Neo-Constructivism, and takes the two empirical scientific experimental results of developmental cognitive neuroscience and spiking neural network as the factual basis, and develops two theories from the perspective of development. Integration, to achieve the resolution of contradictions, complement each other, and then rebuild. This paper discusses Con<span "="">nectivism-Constructivism learning theory. The theory holds that the essence of knowledge is the connecti<span style="letter-spacing:-0.05pt;">on between the subject and the environment. There are two form</span>s: physical form and logical form. </span><span "="">The </span><span "="">only logical form can be realized and utilized by people. Learning can be divided into two stages: connection and construction. Connection is the premise, construction is the core, and the network action generated in the connection stage as a raw material is pruned, and processed by various systems in the construction stage to become psychological representation. When the psychological representation is used, the relevant network shaping is finished, and the meaningful network is formed, which completes the change of knowledge from physical form to logical form and from logical form to physical form. Therefore, learning is the process of constructing meaningful network. We should not only promote the students’ connection stage, but also help the students’ construction stage. The innovation and breakthrough contribution of this paper is that it is the first time to look at the topic of learning theory from a new research perspective. In order t<span style="letter-spacing:-0.05pt;">o explore a more convincing learning theoretical framework, this artic</span>le takes the lead in seeking theoretical support and factual basis from developmental cognitive neuroscience and Spiking neural network. As a result, Connectivism learning theory and Constructivism learning theory are successfully integrated into a rather complete and effective theoretical framework to reconstruct Connectivism-Constructivism learning theory.展开更多
Being progressively applied in the design of highly active catalysts for energy devices,machine learning(ML)technology has shown attractive ability of dramatically reducing the computational cost of the traditional de...Being progressively applied in the design of highly active catalysts for energy devices,machine learning(ML)technology has shown attractive ability of dramatically reducing the computational cost of the traditional density functional theory(DFT)method,showing a particular advantage for the simulation of intricate system catalysis.Starting with a basic description of the whole workflow of the novel DFT-based and ML-accelerated(DFT-ML)scheme,and the common algorithms useable for machine learning,we presented in this paper our work on the development and performance test of a DFT-based ML method for catalysis program(DMCP)to implement the DFT-ML scheme.DMCP is an efficient and user-friendly program with the flexibility to accommodate the needs of performing ML calculations based on the data generated by DFT calculations or from materials database.We also employed an example of transition metal phthalocyanine double-atom catalysts as electrocatalysts for carbon reduction reaction to exhibit the general workflow of the DFT-ML hybrid scheme and our DMCP program.展开更多
In this article, we have developed a game theory based prediction tool, named Preana, based on a promising model developed by Professor Bruce Beuno de Mesquita. The first part of this work is dedicated to exploration ...In this article, we have developed a game theory based prediction tool, named Preana, based on a promising model developed by Professor Bruce Beuno de Mesquita. The first part of this work is dedicated to exploration of the specifics of Mesquita’s algorithm and reproduction of the factors and features that have not been revealed in literature. In addition, we have developed a learning mechanism to model the players’ reasoning ability when it comes to taking risks. Preana can predict the outcome of any issue with multiple steak-holders who have conflicting interests in economic, business, and political sciences. We have utilized game theory, expected utility theory, Median voter theory, probability distribution and reinforcement learning. We were able to reproduce Mesquita’s reported results and have included two case studies from his publications and compared his results to that of Preana. We have also applied Preana on Irans 2013 presidential election to verify the accuracy of the prediction made by Preana.展开更多
基金The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work through Large Groups(Grant Number RGP.2/246/44),B.B.,and https://www.kku.edu.sa/en.
文摘Heart monitoring improves life quality.Electrocardiograms(ECGs or EKGs)detect heart irregularities.Machine learning algorithms can create a few ECG diagnosis processing methods.The first method uses raw ECG and time-series data.The second method classifies the ECG by patient experience.The third technique translates ECG impulses into Q waves,R waves and S waves(QRS)features using richer information.Because ECG signals vary naturally between humans and activities,we will combine the three feature selection methods to improve classification accuracy and diagnosis.Classifications using all three approaches have not been examined till now.Several researchers found that Machine Learning(ML)techniques can improve ECG classification.This study will compare popular machine learning techniques to evaluate ECG features.Four algorithms—Support Vector Machine(SVM),Decision Tree,Naive Bayes,and Neural Network—compare categorization results.SVM plus prior knowledge has the highest accuracy(99%)of the four ML methods.QRS characteristics failed to identify signals without chaos theory.With 99.8%classification accuracy,the Decision Tree technique outperformed all previous experiments.
基金partially supported by the National Natural Science Foundation of China (62173308)the Natural Science Foundation of Zhejiang Province of China (LR20F030001)the Jinhua Science and Technology Project (2022-1-042)。
文摘As a representative emerging machine learning technique, federated learning(FL) has gained considerable popularity for its special feature of “making data available but not visible”. However, potential problems remain, including privacy breaches, imbalances in payment, and inequitable distribution.These shortcomings let devices reluctantly contribute relevant data to, or even refuse to participate in FL. Therefore, in the application of FL, an important but also challenging issue is to motivate as many participants as possible to provide high-quality data to FL. In this paper, we propose an incentive mechanism for FL based on the continuous zero-determinant(CZD) strategies from the perspective of game theory. We first model the interaction between the server and the devices during the FL process as a continuous iterative game. We then apply the CZD strategies for two players and then multiple players to optimize the social welfare of FL, for which we prove that the server can keep social welfare at a high and stable level. Subsequently, we design an incentive mechanism based on the CZD strategies to attract devices to contribute all of their high-accuracy data to FL.Finally, we perform simulations to demonstrate that our proposed CZD-based incentive mechanism can indeed generate high and stable social welfare in FL.
基金financially supported by the National Key Research and Development Program of China(No.2016YFB0701202,No.2017YFB0701500 and No.2020YFB1505901)National Natural Science Foundation of China(General Program No.51474149,52072240)+3 种基金Shanghai Science and Technology Committee(No.18511109300)Science and Technology Commission of the CMC(2019JCJQZD27300)financial support from the University of Michigan and Shanghai Jiao Tong University joint funding,China(AE604401)Science and Technology Commission of Shanghai Municipality(No.18511109302).
文摘Magnesium(Mg)alloys have shown great prospects as both structural and biomedical materials,while poor corrosion resistance limits their further application.In this work,to avoid the time-consuming and laborious experiment trial,a high-throughput computational strategy based on first-principles calculations is designed for screening corrosion-resistant binary Mg alloy with intermetallics,from both the thermodynamic and kinetic perspectives.The stable binary Mg intermetallics with low equilibrium potential difference with respect to the Mg matrix are firstly identified.Then,the hydrogen adsorption energies on the surfaces of these Mg intermetallics are calculated,and the corrosion exchange current density is further calculated by a hydrogen evolution reaction(HER)kinetic model.Several intermetallics,e.g.Y_(3)Mg,Y_(2)Mg and La_(5)Mg,are identified to be promising intermetallics which might effectively hinder the cathodic HER.Furthermore,machine learning(ML)models are developed to predict Mg intermetallics with proper hydrogen adsorption energy employing work function(W_(f))and weighted first ionization energy(WFIE).The generalization of the ML models is tested on five new binary Mg intermetallics with the average root mean square error(RMSE)of 0.11 eV.This study not only predicts some promising binary Mg intermetallics which may suppress the galvanic corrosion,but also provides a high-throughput screening strategy and ML models for the design of corrosion-resistant alloy,which can be extended to ternary Mg alloys or other alloy systems.
基金the financial support from the National Natural Science Foundation of China(22278070,21978047,21776046)。
文摘The high throughput prediction of the thermodynamic phase behavior of active pharmaceutical ingredients(APIs)with pharmaceutically relevant excipients remains a major scientific challenge in the screening of pharmaceutical formulations.In this work,a developed machine-learning model efficiently predicts the solubility of APIs in polymers by learning the phase equilibrium principle and using a few molecular descriptors.Under the few-shot learning framework,thermodynamic theory(perturbed-chain statistical associating fluid theory)was used for data augmentation,and computational chemistry was applied for molecular descriptors'screening.The results showed that the developed machine-learning model can predict the API-polymer phase diagram accurately,broaden the solubility data of APIs in polymers,and reproduce the relationship between API solubility and the interaction mechanisms between API and polymer successfully,which provided efficient guidance for the development of pharmaceutical formulations.
基金This work was supported by the National Natural Science Foundation of China(Grant No.42050104)the Science Foundation of SINOPEC Group(Grant No.P20030).
文摘A comprehensive and precise analysis of shale gas production performance is crucial for evaluating resource potential,designing a field development plan,and making investment decisions.However,quantitative analysis can be challenging because production performance is dominated by the complex interaction among a series of geological and engineering factors.In fact,each factor can be viewed as a player who makes cooperative contributions to the production payoff within the constraints of physical laws and models.Inspired by the idea,we propose a hybrid data-driven analysis framework in this study,where the contributions of dominant factors are quantitatively evaluated,the productions are precisely forecasted,and the development optimization suggestions are comprehensively generated.More specifically,game theory and machine learning models are coupled to determine the dominating geological and engineering factors.The Shapley value with definite physical meaning is employed to quantitatively measure the effects of individual factors.A multi-model-fused stacked model is trained for production forecast,which provides the basis for derivative-free optimization algorithms to optimize the development plan.The complete workflow is validated with actual production data collected from the Fuling shale gas field,Sichuan Basin,China.The validation results show that the proposed procedure can draw rigorous conclusions with quantified evidence and thereby provide specific and reliable suggestions for development plan optimization.Comparing with traditional and experience-based approaches,the hybrid data-driven procedure is advanced in terms of both efficiency and accuracy.
基金Supported by Technology and Innovation Major Project of the Ministry of Science and Technology of China(2020AAA0108400, 2020AAA0108403)Tsinghua Precision Medicine Foundation(10001020109)。
文摘Deep multi-modal learning,a rapidly growing field with a wide range of practical applications,aims to effectively utilize and integrate information from multiple sources,known as modalities.Despite its impressive empirical performance,the theoretical foundations of deep multi-modal learning have yet to be fully explored.In this paper,we will undertake a comprehensive survey of recent developments in multi-modal learning theories,focusing on the fundamental properties that govern this field.Our goal is to provide a thorough collection of current theoretical tools for analyzing multi-modal learning,to clarify their implications for practitioners,and to suggest future directions for the establishment of a solid theoretical foundation for deep multi-modal learning.
文摘In real-time strategy(RTS)games,the ability of recognizing other players’goals is important for creating artifical intelligence(AI)players.However,most current goal recognition methods do not take the player’s deceptive behavior into account which often occurs in RTS game scenarios,resulting in poor recognition results.In order to solve this problem,this paper proposes goal recognition for deceptive agent,which is an extended goal recognition method applying the deductive reason method(from general to special)to model the deceptive agent’s behavioral strategy.First of all,the general deceptive behavior model is proposed to abstract features of deception,and then these features are applied to construct a behavior strategy that best matches the deceiver’s historical behavior data by the inverse reinforcement learning(IRL)method.Final,to interfere with the deceptive behavior implementation,we construct a game model to describe the confrontation scenario and the most effective interference measures.
文摘Attacks on the cyber space is getting exponential in recent times.Illegal penetrations and breaches are real threats to the individuals and organizations.Conventional security systems are good enough to detect the known threats but when it comes to Advanced Persistent Threats(APTs)they fails.These APTs are targeted,more sophisticated and very persistent and incorporates lot of evasive techniques to bypass the existing defenses.Hence,there is a need for an effective defense system that can achieve a complete reliance of security.To address the above-mentioned issues,this paper proposes a novel honeypot system that tracks the anonymous behavior of the APT threats.The key idea of honeypot leverages the concepts of graph theory to detect such targeted attacks.The proposed honey-pot is self-realizing,strategic assisted which withholds the APTs actionable tech-niques and observes the behavior for analysis and modelling.The proposed graph theory based self learning honeypot using the resultsγ(C(n,1)),γc(C(n,1)),γsc(C(n,1))outperforms traditional techniques by detecting APTs behavioral with detection rate of 96%.
文摘This study focuses on the effectiveness of the project-based language learning(PBLL) in a college Secretarial Oral English(SOE) Module. Student reflections of the language project work have been analyzed through Activity Theory. Moreover,Data has been collected and categorized based on the components of complex human activity: the subject, object, tools(signs,symbols, and language), the community in which the activity take place, division of labor, and rules. The findings theoretically support the outcome of project-based language learning which align with the object of the activity.
文摘The unceasing revolution of the global economy and culture boosts the revolutionary step of the educational circle.Combining the creed of The Guide of College English Teaching in 2016 with the results of investigation and survey in colleges, a research group in the Institute of Foreign Languages of Hankou University comes up with a revolutionary trial scheme on College English teaching conducted by discovery learning theory, as well as a research method of action research, which is in hope of mending the problems and shortcomings of current College English teaching.
文摘Since traditional English teaching method, which merely focuses on language teaching but ignores communicative competence, severely impedes the development of students' oral ability. It is high time that English teachers took measures to find a workable and valuable teaching method which can improve students' speaking proficiency effectively. Learning community theory provides a broad space for this, for it regards learning as a process which takes place in a community where the learners are sharing their experience towards knowledge building in an interactive and cooperative way.
文摘This paper tries to summarize some main schools 0f teaching methodologies abroad and some main learning theories abroad. From this paper, we can know the main learning theories, the basic theories of them and the leading figures. It can help us understand the characteristics of each school of the teaching methodologies and learning theories.
基金supported by the National Natural Science Founda-tion of China(No.22203026,No.22203025,and No.12174080)the National Key R&D Program of China(No.2022YFA1602601)+1 种基金the Fundamental Research Funds for the Central Universities(JZ2022HGTA0313 and JZ2022HGQA0198)the Anhui Provincial Nat-ural Science Foundation(2208085QB44).
文摘Lithium has been paid great attention in recent years thanks to its significant appli-cations for battery and lightweight alloy.Developing a potential model with high ac-curacy and efficiency is impor-tant for theoretical simulation of lithium materials.Here,we build a deep learning potential(DP)for elemental lithium based on a concurrent-learning scheme and DP representation of the density-functional theory(DFT)potential energy surface(PES),the DP model enables material simulations with close-to DFT accuracy but at much lower computational cost.The simulations show that basic parameters,equation of states,elasticity,defects and surface are consistent with the first principles results.More notably,the liquid radial distribution func-tion based on our DP model is found to match well with experiment data.Our results demon-strate that the developed DP model can be used for the simulation of lithium materials.
基金funded by the Beijing Natural Science Foundation (4202002).
文摘Social engineering attacks are considered one of the most hazardous cyberattacks in cybersecurity,as human vulnerabilities are often the weakest link in the entire network.Such vulnerabilities are becoming increasingly susceptible to network security risks.Addressing the social engineering attack defense problem has been the focus of many studies.However,two main challenges hinder its successful resolution.Firstly,the vulnerabilities in social engineering attacks are unique due to multistage attacks,leading to incorrect social engineering defense strategies.Secondly,social engineering attacks are real-time,and the defense strategy algorithms based on gaming or reinforcement learning are too complex to make rapid decisions.This paper proposes a multiattribute quantitative incentive method based on human vulnerability and an improved Q-learning(IQL)reinforcement learning method on human vulnerability attributes.The proposed algorithm aims to address the two main challenges in social engineering attack defense by using a multiattribute incentive method based on human vulnerability to determine the optimal defense strategy.Furthermore,the IQL reinforcement learning method facilitates rapid decision-making during real-time attacks.The experimental results demonstrate that the proposed algorithm outperforms the traditional Qlearning(QL)and deep Q-network(DQN)approaches in terms of time efficiency,taking 9.1%and 19.4%less time,respectively.Moreover,the proposed algorithm effectively addresses the non-uniformity of vulnerabilities in social engineering attacks and provides a reliable defense strategy based on human vulnerability attributes.This study contributes to advancing social engineering attack defense by introducing an effective and efficient method for addressing the vulnerabilities of human factors in the cybersecurity domain.
文摘Second language acquisition can not be understood without addressing the interaction between language and cognition. Cognitive theory can extend to describe learning strategies as complex cognitive skills. Theoretical developments in Anderson’s production systems cover a broader range of behavior than other theories, including comprehension and production of oral and written texts as well as comprehension, problem solving, and verbal learning.Thus Anderson’s cognitive theory can be served as a rationale for learning strategy studies in second language acquisition.
文摘By analyzing the English learning logs of 12 students in a provincial university in south-west China after they had been exempted from taking college English courses,this study investigated college students’autonomous EFL(English as a foreign language)learning after course exemption,including the use of mediational means in EFL learning,EFL learning hours,and other factors affecting EFL learning,in the hope of giving new perspectives on college ELF curriculum design,teaching,and education management.
文摘This systematic review is focused on the importance of the Bandura Social Cognitive Theory and its theoretical components such as self-efficacy in workplace of today.The themes have been described through the course of the decades.These theories have been utilized heavily in research and real-life case studies,have been further developed by Bandura and other researchers,and have been implemented in organizational psychology.During the past 10 years they have helped in reshaping Human Resources Development.Major latest contributions and applications are discussed touching even the recent outbreak of the pandemic.The influence of the theory is immense and the importance of self-efficacy in the workplace has been addressed and proven by research.
文摘Based on a new theoretical perspective, this paper attempts to unify the seemingly incompatible learning theories of Connectivism and Constructivism into a scientific theoretical framework. The Connectivism-Constructivism learning theory is not a simple superposition of the two theories. Instead, it absorbs the essence of the learning theory of Constructivism, Connectivism and Neo-Constructivism, and takes the two empirical scientific experimental results of developmental cognitive neuroscience and spiking neural network as the factual basis, and develops two theories from the perspective of development. Integration, to achieve the resolution of contradictions, complement each other, and then rebuild. This paper discusses Con<span "="">nectivism-Constructivism learning theory. The theory holds that the essence of knowledge is the connecti<span style="letter-spacing:-0.05pt;">on between the subject and the environment. There are two form</span>s: physical form and logical form. </span><span "="">The </span><span "="">only logical form can be realized and utilized by people. Learning can be divided into two stages: connection and construction. Connection is the premise, construction is the core, and the network action generated in the connection stage as a raw material is pruned, and processed by various systems in the construction stage to become psychological representation. When the psychological representation is used, the relevant network shaping is finished, and the meaningful network is formed, which completes the change of knowledge from physical form to logical form and from logical form to physical form. Therefore, learning is the process of constructing meaningful network. We should not only promote the students’ connection stage, but also help the students’ construction stage. The innovation and breakthrough contribution of this paper is that it is the first time to look at the topic of learning theory from a new research perspective. In order t<span style="letter-spacing:-0.05pt;">o explore a more convincing learning theoretical framework, this artic</span>le takes the lead in seeking theoretical support and factual basis from developmental cognitive neuroscience and Spiking neural network. As a result, Connectivism learning theory and Constructivism learning theory are successfully integrated into a rather complete and effective theoretical framework to reconstruct Connectivism-Constructivism learning theory.
文摘Being progressively applied in the design of highly active catalysts for energy devices,machine learning(ML)technology has shown attractive ability of dramatically reducing the computational cost of the traditional density functional theory(DFT)method,showing a particular advantage for the simulation of intricate system catalysis.Starting with a basic description of the whole workflow of the novel DFT-based and ML-accelerated(DFT-ML)scheme,and the common algorithms useable for machine learning,we presented in this paper our work on the development and performance test of a DFT-based ML method for catalysis program(DMCP)to implement the DFT-ML scheme.DMCP is an efficient and user-friendly program with the flexibility to accommodate the needs of performing ML calculations based on the data generated by DFT calculations or from materials database.We also employed an example of transition metal phthalocyanine double-atom catalysts as electrocatalysts for carbon reduction reaction to exhibit the general workflow of the DFT-ML hybrid scheme and our DMCP program.
文摘In this article, we have developed a game theory based prediction tool, named Preana, based on a promising model developed by Professor Bruce Beuno de Mesquita. The first part of this work is dedicated to exploration of the specifics of Mesquita’s algorithm and reproduction of the factors and features that have not been revealed in literature. In addition, we have developed a learning mechanism to model the players’ reasoning ability when it comes to taking risks. Preana can predict the outcome of any issue with multiple steak-holders who have conflicting interests in economic, business, and political sciences. We have utilized game theory, expected utility theory, Median voter theory, probability distribution and reinforcement learning. We were able to reproduce Mesquita’s reported results and have included two case studies from his publications and compared his results to that of Preana. We have also applied Preana on Irans 2013 presidential election to verify the accuracy of the prediction made by Preana.