Task allocation for munition swarms is constrained by reachable region limitations and real-time requirements.This paper proposes a reachable region guided distributed coalition formation game(RRGDCF)method to address...Task allocation for munition swarms is constrained by reachable region limitations and real-time requirements.This paper proposes a reachable region guided distributed coalition formation game(RRGDCF)method to address these issues.To enable efficient online task allocation,a reachable region prediction strategy based on fully connected neural networks(FCNNs)is developed.This strategy integrates high-fidelity data generated from the golden section method and low-fidelity data from geometric approximation in an optimal mixing ratio to form multi-fidelity samples,significantly enhancing prediction accuracy and efficiency under limited high-fidelity samples.These predictions are then incorporated into the coalition formation game framework.A tabu search mechanism guided by the reachable region center directs munitions to execute tasks within their respective reachable regions,mitigating redundant operations on ineffective coalition structures.Furthermore,an adaptive guidance coalition formation strategy optimizes allocation plans by leveraging the hit probabilities of munitions,replacing traditional random coalition formation methods.Simulation results demonstrate that RRGDCF surpasses the contract network protocol and traditional coalition formation game algorithms in optimality and computational efficiency.Hardware experiments further validate the method's practicality in dynamic scenarios.展开更多
The emergence of multi-access edge computing(MEC)aims at extending cloud computing capabilities to the edge of the radio access network.As the large-scale internet of things(IoT)services are rapidly growing,a single e...The emergence of multi-access edge computing(MEC)aims at extending cloud computing capabilities to the edge of the radio access network.As the large-scale internet of things(IoT)services are rapidly growing,a single edge infrastructure provider(EIP)may not be sufficient to handle the data traffic generated by these services.Most of the existing work addressed the computing resource shortage problem by optimizing tasks schedule,whereas others overcome such issue by placing computing resources on demand.However,when considering a multiple EIPs scenario,an urgent challenge is how to generate a coalition structure to maximize each EIP’s gain with a suitable price for computing resource block corresponding to a container.To this end,we design a scheme of EIPs collaboration with a market price for containers under a scenario that considers a collection of service providers(SPs)with different budgets and several EIPs distributed in geographical locations.First,we bring in the net profit market price model to generate a more reasonable equilibrium price and select the optimal EIPs for each SP by a convex program.Then we use a mathematical model to maximize EIP’s profits and form stable coalitions between EIPs by a distributed coalition formation algorithm.Numerical results demonstrate that our proposed collaborative scheme among EIPs enhances EIPs’gain and increases users’surplus.展开更多
Underground engineering projects such as deep tunnel excavation often encounter rockburst disasters accompanied by numerous microseismic events.Rapid interpretation of microseismic signals is crucial for the timely id...Underground engineering projects such as deep tunnel excavation often encounter rockburst disasters accompanied by numerous microseismic events.Rapid interpretation of microseismic signals is crucial for the timely identification of rockbursts.However,conventional processing encompasses multi-step workflows,including classification,denoising,picking,locating,and computational analysis,coupled with manual intervention,which collectively compromise the reliability of early warnings.To address these challenges,this study innovatively proposes the“microseismic stethoscope"-a multi-task machine learning and deep learning model designed for the automated processing of massive microseismic signals.This model efficiently extracts three key parameters that are necessary for recognizing rockburst disasters:rupture location,microseismic energy,and moment magnitude.Specifically,the model extracts raw waveform features from three dedicated sub-networks:a classifier for source zone classification,and two regressors for microseismic energy and moment magnitude estimation.This model demonstrates superior efficiency compared to traditional processing and semi-automated processing,reducing per-event processing time from 0.71 s to 0.49 s to merely 0.036 s.It concurrently achieves 98%accuracy in source zone classification,with microseismic energy and moment magnitude estimation errors of 0.13 and 0.05,respectively.This model has been well applied and validated in the Daxiagu Tunnel case in Sichuan,China.The application results indicate that the model is as accurate as traditional methods in determining source parameters,and thus can be used to identify potential geomechanical processes of rockburst disasters.By enhancing the signal processing reliability of microseismic events,the proposed model in this study presents a significant advancement in the identification of rockburst disasters.展开更多
Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Re...Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.展开更多
Surface properties of crystals are critical in many fields,including electrochemistry and photoelectronics,the efficient prediction of which can expedite the design and optimization of catalysts,batteries,alloys etc.H...Surface properties of crystals are critical in many fields,including electrochemistry and photoelectronics,the efficient prediction of which can expedite the design and optimization of catalysts,batteries,alloys etc.However,we are still far from realizing this vision due to the rarity of surface property-related databases,especially for multicomponent compounds,due to the large sample spaces and limited computing resources.In this work,we present a surface emphasized multi-task crystal graph convolutional neural network(SEM-CGCNN)to predict multiple surface properties simultaneously from crystal structures.The model is evaluated on a dataset of 3526 surface energies and work functions of binary magnesium intermetallics obtained through first-principles calculations,and obvious improvements are observed both in efficiency and accuracy over the original CGCNN model.By transferring the pre-trained model to the datasets of pure metals and other intermetallics,the fine-tuned SEM-CGCNN outperforms learning from scratch and can be further applied to other surface properties and materials systems.This study could be a paradigm for the end-to-end mapping of atomic structures to anisotropic surface properties of crystals,which provides an efficient framework to understand and screen materials with desired surface characteristics.展开更多
Reconfigurable intelligent surface(RIS)have been cast as a promising alternative to alleviate blockage vulnerability and enhance coverage capability for terahertz(THz)communications.Owing to large-scale array elements...Reconfigurable intelligent surface(RIS)have been cast as a promising alternative to alleviate blockage vulnerability and enhance coverage capability for terahertz(THz)communications.Owing to large-scale array elements at transceivers and RIS,the codebook based beamforming can be utilized in a computationally efficient manner.However,the codeword selection for analog beamforming is an intractable combinatorial optimization(CO)problem.To this end,by taking the CO problem as a classification problem,a multi-task learning based analog beam selection(MTL-ABS)framework is developed to implement cooperative beam selection concurrently at transceivers and RIS.In addition,residual network and self-attention mechanism are used to combat the network degradation and mine intrinsic THz channel features.Finally,the network convergence is analyzed from a blockwise perspective,and numerical results demonstrate that the MTL-ABS framework greatly decreases the beam selection overhead and achieves near optimal sum-rate compared with heuristic search based counterparts.展开更多
The formation of the manned aerial vehicle/unmanned aerial vehicle(MAV/UAV) task coalition is considered. To reduce the scale of the problem, the formation progress is divided into three phases. For the task clusterin...The formation of the manned aerial vehicle/unmanned aerial vehicle(MAV/UAV) task coalition is considered. To reduce the scale of the problem, the formation progress is divided into three phases. For the task clustering phase, the geographical position of tasks is taken into consideration and a cluster method is proposed. For the UAV allocation phase, the UAV requirement for both constrained and unconstrained resources is introduced, and a multi-objective optimal algorithm is proposed to solve the allocation problem. For the MAV allocation phase, the optimal model is firstly constructed and it is decomposed according to the ideal of greed to reduce the time complexity of the algorithm. Based on the above phases, the MAV/UAV task coalition formation method is proposed and the effectiveness and practicability are demonstrated by simulation examples.展开更多
This study analyzes the cooperative coalition problem for formation scheduling based on incomplete information. A multi-agent cooperative coalition framework is developed to optimize the formation scheduling problem i...This study analyzes the cooperative coalition problem for formation scheduling based on incomplete information. A multi-agent cooperative coalition framework is developed to optimize the formation scheduling problem in a decentralized manner. The social class differentiation mech- anism and role-assuming mechanism are incorporated into the framework, which, in turn, ensures that the multi-agent system （MAS） evolves in the optimal direction. Moreover, a further differen- tiation pressure can be achieved to help MAS escape from local optima. A Bayesian coalition nego- tiation algorithm is constructed, within which the Harsanyi transformation is introduced to transform the coalition problem based on incomplete information to the Bayesian-equivalent coali- tion problem based on imperfect information. The simulation results suggest that the distribution of agents＇ expectations of other agents＇ unknown information approximates to the true distribution after a finite set of generations. The comparisons indicate that the MAS cooperative coalition algo- rithm produces a significantly better utility and possesses a more effective capability of escaping from local optima than the proposal-engaged marriage algorithm and the Simulated Annealing algorithm.展开更多
A coalition formation algorithm is presented with limited communication ranges and delays in unknown environment,for the performance of multiple heterogeneous unmanned aerial vehicles(UAVs)in cooperative search and at...A coalition formation algorithm is presented with limited communication ranges and delays in unknown environment,for the performance of multiple heterogeneous unmanned aerial vehicles(UAVs)in cooperative search and attack missions.The mathematic model of coalition formation is built on basis of the minimum attacking time and the minimum coalition size with satisfying resources and simultaneous strikes requirements.A communication protocol based on maximum number of hops is developed to determine the potential coalition members in dynamic network.A multistage sub-optimal coalition formation algorithm(MSOCFA)with polynomial time is established.The performances of MSOCFA and particle swarm optimization(PSO)algorithms are compared in terms of complexity,mission performance and computational time.A complex scenario is deployed to illustrate how the coalitions are formed and validate the feasibility of the MSOCFA.The effect of communication constraints(hop delay and max-hops)on mission performance is studied.The results show that it is beneficial to determine potential coalition members in a wide and deep range over the network in the presence of less delay.However,when the delays are significant,it is more advantageous to determine coalitions from among the immediate neighbors.展开更多
In multi-agent systems, autonomous agents may form coalition to increase the efficiency of problem solving. But the current coalition algorithm is very complex, and cannot satisfy the condition of optimality and stabl...In multi-agent systems, autonomous agents may form coalition to increase the efficiency of problem solving. But the current coalition algorithm is very complex, and cannot satisfy the condition of optimality and stableness simultaneously. To solve the problem, an algorithm that uses the mechanism of distribution according to work for coalition formation is presented, which can achieve global optimal and stable solution in subadditive task oriented domains. The validity of the algorithm is demonstrated by both experiments and theory.展开更多
For the issue of evaluation of capability of enterprise agent coalition,an evaluation model based on information fusion and entropy weighting method is presented. The attribute reduction method is utilized to reduce i...For the issue of evaluation of capability of enterprise agent coalition,an evaluation model based on information fusion and entropy weighting method is presented. The attribute reduction method is utilized to reduce indicators of the capability according to the theory of rough set. The new indicator system can be determined. Attribute reduction can also reduce the workload and remove the redundant information,when there are too many indicators or the indicators have strong correlation. The research complexity can be reduced and the efficiency can be improved. Entropy weighting method is used to determine the weights of the remaining indicators,and the importance of indicators is analyzed. The information fusion model based on nearest neighbor method is developed and utilized to evaluate the capability of multiple agent coalitions,compared to cloud evaluation model and D-S evidence method. Simulation results are reasonable and with obvious distinction. Thus they verify the effectiveness and feasibility of the model. The information fusion model can provide more scientific,rational decision support for choosing the best agent coalition,and provide innovative steps for the evaluation process of capability of agent coalitions.展开更多
In this paper, we introduce a simple coalition formation game in the environment of bidding, which is a special case of the weighted majority game （WMG）, and is named the weighted simple-majority game （WSMG）. In W...In this paper, we introduce a simple coalition formation game in the environment of bidding, which is a special case of the weighted majority game （WMG）, and is named the weighted simple-majority game （WSMG）. In WSMG, payoff is allocated to the winners proportional to the players powers, which can be measured in various ways. We define a new kind of stability： the counteraction-stability （C-stability）, where any potential deviating players will confront counteractions of the other players. We show that C-stable coalition structures in WSMG always contains a minimal winning coalition of minimum total power. For the variant where powers are measured directly by their weights, we show that it is NP-hard to find a C-stable coalition structure and design a pseudo-polynomial time algorithm. Sensitivity analysis for this variant, which shows many interesting properties, is also done. We also prove that it is NP-hard to compute the Holler-Packel indices in WSMGs, and hence in WMGs as well.展开更多
In this paper,a generalized form of the symmetric Banzhaf value for cooperative fuzzy games with a coalition structure is proposed.Three axiomatic systems of the symmetric Banzhaf value are given by extending crisp ca...In this paper,a generalized form of the symmetric Banzhaf value for cooperative fuzzy games with a coalition structure is proposed.Three axiomatic systems of the symmetric Banzhaf value are given by extending crisp case.Furthermore,we study the symmetric Banzhaf values for two special kinds of fuzzy games,which are called fuzzy games with multilinear extension form and a coalition structure,and fuzzy games with Choquet integral form and a coalition structure,respectively.展开更多
In cooperative game theory, a central problem is to allocate fairly the win of the grand coalition to the players who agreed to cooperate and form the grand coalition. Such allocations are obtained by means of values,...In cooperative game theory, a central problem is to allocate fairly the win of the grand coalition to the players who agreed to cooperate and form the grand coalition. Such allocations are obtained by means of values, having some fairness properties, expressed in most cases by groups of axioms. In an earlier work, we solved what we called the Inverse Problem for Semivalues, in which the main result was offering an explicit formula providing the set of all games with an a priori given Semivalue, associated with a given weight vector. However, in this set there is an infinite set of games for which the Semivalues are not coalitional rational, perhaps not efficient, so that these are not fair practical solutions of the above fundamental problem. Among the Semivalues, coalitional rational solutions for the Shapley Value and the Banzhaf Value have been given in two more recent works. In the present paper, based upon a general potential basis, relative to Semivalues, for a given game and a given Semivalue, we solve the connected problem: in the Inverse Set, find out a game with the same Semivalue, which is also coalitional rational. Several examples will illustrate the corresponding numerical technique.展开更多
Wireless sensor networks(WSNs)are the major contributors to big data acquisition.The authenticity and integrity of the data are two most important basic requirements for various services based on big data.Data aggrega...Wireless sensor networks(WSNs)are the major contributors to big data acquisition.The authenticity and integrity of the data are two most important basic requirements for various services based on big data.Data aggregation is a promising method to decrease operation cost for resource-constrained WSNs.However,the process of data acquisitions in WSNs are in open environments,data aggregation is vulnerable to more special security attacks with hiding feature and subjective fraudulence,such as coalition attack.Aimed to provide data authenticity and integrity protection for WSNs,an efficient and secure identity-based aggregate signature scheme(EIAS)is proposed in this paper.Rigorous security proof shows that our proposed scheme can be secure against all kinds of attacks.The performance comparisons shows EIAS has clear advantages in term of computation cost and communication cost when compared with similar data aggregation scheme for WSNs.展开更多
This article aims to explore the coalition of external actors and the strategies it deployed to influence the emergence of the National Nutrition Policy (NNP) in Lao People’s Democratic Republic (Lao PDR). The Advoca...This article aims to explore the coalition of external actors and the strategies it deployed to influence the emergence of the National Nutrition Policy (NNP) in Lao People’s Democratic Republic (Lao PDR). The Advocacy Coalition Framework and the conceptual model of Effective Advocacy Strategies for Influencing Government Nutrition Policy were used to frame the data collection and their analysis. Sources of information were semi-structured interviews conducted with government and external actors, as well as all available documents on nutrition policy in Laos. The commitment of the government to achieve the Millennium Development Goals (MDGs) and to leave the Least Developed Country status created a favorable condition to support the emergence of the NNP in Laos. This context was a driving force for the building of an effective and convincing coalition of United Nations agencies able to accompany the government in redefining health priorities. Various strategies were used by the coalition to this end, including generating, disseminating, and using scientific evidence, assisting the government with a budget and technical expertise, providing decision-makers with opportunities to learn from other countries, and building relationships with the key actor. External actors can be a major force to support the emergence of a public policy in Laos, but this requires a window of opportunity like what the MDGs have been able to bring.展开更多
The effective classification of urban domestic waste is the key to achieve a “waste-free city” and provides an essential guarantee for resource utilization. This article takes a coalitional game perspective to study...The effective classification of urban domestic waste is the key to achieve a “waste-free city” and provides an essential guarantee for resource utilization. This article takes a coalitional game perspective to study the dilemmas in urban domestic waste separation from the cooperative interaction of residents, government, and enterprises. The study finds that urban domestic waste classification in China is currently facing many problems, focusing on: 1) insufficient consensus among residents, 2) shortage of input funds, 3) corporate profitability difficulties, 4) weak policy constraints, and 5) difficulties in integrating goals. In this regard, each participating body still needs to focus on collective interests, coalitional games, break the dilemma society, and promote the long-term management of urban domestic waste.展开更多
Coalition game theory is introduced to investigate the performance,fairness and stability of decorrelating group multiuser detection receiver,not only from the perspective of individual nodes,but also various coalitio...Coalition game theory is introduced to investigate the performance,fairness and stability of decorrelating group multiuser detection receiver,not only from the perspective of individual nodes,but also various coalitions and the whole system as well. Firstly,to derive how the system scale with coalition size,a stochastic model with transferable payoffs (stochastic TU-model) is provided. Secondly,to find the most preferred coalition structures from the view point of individual nodes,a model with Non-Transferable payoffs (NTU-model) is presented. Theoretical analysis and simulation results suggest that stochasticaly the grand coalition is payoff maximizing for the system as a whole,while individual nodes with good-conditioned channels may prefer local "win-win coalitions".展开更多
This paper proposes a multi-objective optimization design method based on the coalition cooperative game theory where the three design goals have been seen as three game players. By calculating the affecting factors a...This paper proposes a multi-objective optimization design method based on the coalition cooperative game theory where the three design goals have been seen as three game players. By calculating the affecting factors and fuzzy clustering, the design variables are divided into different strategic spaces which belong to each player, then it constructs a payoff function based on the coalition mechanism. Each game player takes its own revenue function as a target and obtains the best strategy versus other players. The best strategies of all players consist of the strategy permutation of a round game and it obtains the final game solutions through multi-round games according to the convergence criterion. A multi-objective optimization example of the luff mechanism of compensative sheave block shows the effectiveness of the coalition cooperative game method.展开更多
基金Supported by National Natural Science Foundation of China（60474035）,National Research Foundation for the Doctoral Program of Higher Education of China（20050359004）,Natural Science Foundation of Anhui Province（070412035）
基金supported by the National Natural Science Foundation of China(Grant 52372347,52425211,52272360)。
文摘Task allocation for munition swarms is constrained by reachable region limitations and real-time requirements.This paper proposes a reachable region guided distributed coalition formation game(RRGDCF)method to address these issues.To enable efficient online task allocation,a reachable region prediction strategy based on fully connected neural networks(FCNNs)is developed.This strategy integrates high-fidelity data generated from the golden section method and low-fidelity data from geometric approximation in an optimal mixing ratio to form multi-fidelity samples,significantly enhancing prediction accuracy and efficiency under limited high-fidelity samples.These predictions are then incorporated into the coalition formation game framework.A tabu search mechanism guided by the reachable region center directs munitions to execute tasks within their respective reachable regions,mitigating redundant operations on ineffective coalition structures.Furthermore,an adaptive guidance coalition formation strategy optimizes allocation plans by leveraging the hit probabilities of munitions,replacing traditional random coalition formation methods.Simulation results demonstrate that RRGDCF surpasses the contract network protocol and traditional coalition formation game algorithms in optimality and computational efficiency.Hardware experiments further validate the method's practicality in dynamic scenarios.
基金supported by National Natural Science Foundation of China(No.6206020135)Key Research and Development Program of Gansu Province(No.20YF8GA123)+1 种基金Gansu Provincial Department of Education University Faculty Innovation Fund Project(No.2024B-059)Youth Science Fund Project of Lanzhou Jiaotong University(No.1200061307).
文摘The emergence of multi-access edge computing(MEC)aims at extending cloud computing capabilities to the edge of the radio access network.As the large-scale internet of things(IoT)services are rapidly growing,a single edge infrastructure provider(EIP)may not be sufficient to handle the data traffic generated by these services.Most of the existing work addressed the computing resource shortage problem by optimizing tasks schedule,whereas others overcome such issue by placing computing resources on demand.However,when considering a multiple EIPs scenario,an urgent challenge is how to generate a coalition structure to maximize each EIP’s gain with a suitable price for computing resource block corresponding to a container.To this end,we design a scheme of EIPs collaboration with a market price for containers under a scenario that considers a collection of service providers(SPs)with different budgets and several EIPs distributed in geographical locations.First,we bring in the net profit market price model to generate a more reasonable equilibrium price and select the optimal EIPs for each SP by a convex program.Then we use a mathematical model to maximize EIP’s profits and form stable coalitions between EIPs by a distributed coalition formation algorithm.Numerical results demonstrate that our proposed collaborative scheme among EIPs enhances EIPs’gain and increases users’surplus.
基金supported by the National Natural Science Foundation of China(Grant Nos.42130719 and 42177173)the Doctoral Direct Train Project of Chongqing Natural Science Foundation(Grant No.CSTB2023NSCQ-BSX0029).
文摘Underground engineering projects such as deep tunnel excavation often encounter rockburst disasters accompanied by numerous microseismic events.Rapid interpretation of microseismic signals is crucial for the timely identification of rockbursts.However,conventional processing encompasses multi-step workflows,including classification,denoising,picking,locating,and computational analysis,coupled with manual intervention,which collectively compromise the reliability of early warnings.To address these challenges,this study innovatively proposes the“microseismic stethoscope"-a multi-task machine learning and deep learning model designed for the automated processing of massive microseismic signals.This model efficiently extracts three key parameters that are necessary for recognizing rockburst disasters:rupture location,microseismic energy,and moment magnitude.Specifically,the model extracts raw waveform features from three dedicated sub-networks:a classifier for source zone classification,and two regressors for microseismic energy and moment magnitude estimation.This model demonstrates superior efficiency compared to traditional processing and semi-automated processing,reducing per-event processing time from 0.71 s to 0.49 s to merely 0.036 s.It concurrently achieves 98%accuracy in source zone classification,with microseismic energy and moment magnitude estimation errors of 0.13 and 0.05,respectively.This model has been well applied and validated in the Daxiagu Tunnel case in Sichuan,China.The application results indicate that the model is as accurate as traditional methods in determining source parameters,and thus can be used to identify potential geomechanical processes of rockburst disasters.By enhancing the signal processing reliability of microseismic events,the proposed model in this study presents a significant advancement in the identification of rockburst disasters.
文摘Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.
基金supported by the National Key R&D Program(No.2021YFB3501002)supported by the Ministry of Science and Technology of China,National Natural Science Foundation of China(No.51825101,52127801).
文摘Surface properties of crystals are critical in many fields,including electrochemistry and photoelectronics,the efficient prediction of which can expedite the design and optimization of catalysts,batteries,alloys etc.However,we are still far from realizing this vision due to the rarity of surface property-related databases,especially for multicomponent compounds,due to the large sample spaces and limited computing resources.In this work,we present a surface emphasized multi-task crystal graph convolutional neural network(SEM-CGCNN)to predict multiple surface properties simultaneously from crystal structures.The model is evaluated on a dataset of 3526 surface energies and work functions of binary magnesium intermetallics obtained through first-principles calculations,and obvious improvements are observed both in efficiency and accuracy over the original CGCNN model.By transferring the pre-trained model to the datasets of pure metals and other intermetallics,the fine-tuned SEM-CGCNN outperforms learning from scratch and can be further applied to other surface properties and materials systems.This study could be a paradigm for the end-to-end mapping of atomic structures to anisotropic surface properties of crystals,which provides an efficient framework to understand and screen materials with desired surface characteristics.
文摘Reconfigurable intelligent surface(RIS)have been cast as a promising alternative to alleviate blockage vulnerability and enhance coverage capability for terahertz(THz)communications.Owing to large-scale array elements at transceivers and RIS,the codebook based beamforming can be utilized in a computationally efficient manner.However,the codeword selection for analog beamforming is an intractable combinatorial optimization(CO)problem.To this end,by taking the CO problem as a classification problem,a multi-task learning based analog beam selection(MTL-ABS)framework is developed to implement cooperative beam selection concurrently at transceivers and RIS.In addition,residual network and self-attention mechanism are used to combat the network degradation and mine intrinsic THz channel features.Finally,the network convergence is analyzed from a blockwise perspective,and numerical results demonstrate that the MTL-ABS framework greatly decreases the beam selection overhead and achieves near optimal sum-rate compared with heuristic search based counterparts.
基金supported by the National Natural Science Foundation of China(61573017 61703425)the Aeronautical Science Fund(20175796014)
文摘The formation of the manned aerial vehicle/unmanned aerial vehicle(MAV/UAV) task coalition is considered. To reduce the scale of the problem, the formation progress is divided into three phases. For the task clustering phase, the geographical position of tasks is taken into consideration and a cluster method is proposed. For the UAV allocation phase, the UAV requirement for both constrained and unconstrained resources is introduced, and a multi-objective optimal algorithm is proposed to solve the allocation problem. For the MAV allocation phase, the optimal model is firstly constructed and it is decomposed according to the ideal of greed to reduce the time complexity of the algorithm. Based on the above phases, the MAV/UAV task coalition formation method is proposed and the effectiveness and practicability are demonstrated by simulation examples.
基金supported by the National Natural Science Foundation of China(No.61039001)the National Science and Technology Support Program of China(No.2011BAH24B10)
文摘This study analyzes the cooperative coalition problem for formation scheduling based on incomplete information. A multi-agent cooperative coalition framework is developed to optimize the formation scheduling problem in a decentralized manner. The social class differentiation mech- anism and role-assuming mechanism are incorporated into the framework, which, in turn, ensures that the multi-agent system （MAS） evolves in the optimal direction. Moreover, a further differen- tiation pressure can be achieved to help MAS escape from local optima. A Bayesian coalition nego- tiation algorithm is constructed, within which the Harsanyi transformation is introduced to transform the coalition problem based on incomplete information to the Bayesian-equivalent coali- tion problem based on imperfect information. The simulation results suggest that the distribution of agents＇ expectations of other agents＇ unknown information approximates to the true distribution after a finite set of generations. The comparisons indicate that the MAS cooperative coalition algo- rithm produces a significantly better utility and possesses a more effective capability of escaping from local optima than the proposal-engaged marriage algorithm and the Simulated Annealing algorithm.
基金partially sponsored by the Fundamental Research Funds for the Central Universities(No.3102015ZY092)
文摘A coalition formation algorithm is presented with limited communication ranges and delays in unknown environment,for the performance of multiple heterogeneous unmanned aerial vehicles(UAVs)in cooperative search and attack missions.The mathematic model of coalition formation is built on basis of the minimum attacking time and the minimum coalition size with satisfying resources and simultaneous strikes requirements.A communication protocol based on maximum number of hops is developed to determine the potential coalition members in dynamic network.A multistage sub-optimal coalition formation algorithm(MSOCFA)with polynomial time is established.The performances of MSOCFA and particle swarm optimization(PSO)algorithms are compared in terms of complexity,mission performance and computational time.A complex scenario is deployed to illustrate how the coalitions are formed and validate the feasibility of the MSOCFA.The effect of communication constraints(hop delay and max-hops)on mission performance is studied.The results show that it is beneficial to determine potential coalition members in a wide and deep range over the network in the presence of less delay.However,when the delays are significant,it is more advantageous to determine coalitions from among the immediate neighbors.
文摘In multi-agent systems, autonomous agents may form coalition to increase the efficiency of problem solving. But the current coalition algorithm is very complex, and cannot satisfy the condition of optimality and stableness simultaneously. To solve the problem, an algorithm that uses the mechanism of distribution according to work for coalition formation is presented, which can achieve global optimal and stable solution in subadditive task oriented domains. The validity of the algorithm is demonstrated by both experiments and theory.
基金Sponsored by the National Natural Science Foundation of China(Grant No.61173052)the China Postdoctoral Scinece Foundation(Grant No.2014M561363)
文摘For the issue of evaluation of capability of enterprise agent coalition,an evaluation model based on information fusion and entropy weighting method is presented. The attribute reduction method is utilized to reduce indicators of the capability according to the theory of rough set. The new indicator system can be determined. Attribute reduction can also reduce the workload and remove the redundant information,when there are too many indicators or the indicators have strong correlation. The research complexity can be reduced and the efficiency can be improved. Entropy weighting method is used to determine the weights of the remaining indicators,and the importance of indicators is analyzed. The information fusion model based on nearest neighbor method is developed and utilized to evaluate the capability of multiple agent coalitions,compared to cloud evaluation model and D-S evidence method. Simulation results are reasonable and with obvious distinction. Thus they verify the effectiveness and feasibility of the model. The information fusion model can provide more scientific,rational decision support for choosing the best agent coalition,and provide innovative steps for the evaluation process of capability of agent coalitions.
基金supported by National Natural Science Foundationof China(No. 70425004)
文摘In this paper, we introduce a simple coalition formation game in the environment of bidding, which is a special case of the weighted majority game （WMG）, and is named the weighted simple-majority game （WSMG）. In WSMG, payoff is allocated to the winners proportional to the players powers, which can be measured in various ways. We define a new kind of stability： the counteraction-stability （C-stability）, where any potential deviating players will confront counteractions of the other players. We show that C-stable coalition structures in WSMG always contains a minimal winning coalition of minimum total power. For the variant where powers are measured directly by their weights, we show that it is NP-hard to find a C-stable coalition structure and design a pseudo-polynomial time algorithm. Sensitivity analysis for this variant, which shows many interesting properties, is also done. We also prove that it is NP-hard to compute the Holler-Packel indices in WSMGs, and hence in WMGs as well.
基金supported by Natural Science Foundation Youth Project of China(No.71201089)National Natural Science Foundation of China(Nos.71071018 and 71271217)Natural Science Foundation Youth Project of Shandong Province,China(No.ZR2012GQ005)
文摘In this paper,a generalized form of the symmetric Banzhaf value for cooperative fuzzy games with a coalition structure is proposed.Three axiomatic systems of the symmetric Banzhaf value are given by extending crisp case.Furthermore,we study the symmetric Banzhaf values for two special kinds of fuzzy games,which are called fuzzy games with multilinear extension form and a coalition structure,and fuzzy games with Choquet integral form and a coalition structure,respectively.
文摘In cooperative game theory, a central problem is to allocate fairly the win of the grand coalition to the players who agreed to cooperate and form the grand coalition. Such allocations are obtained by means of values, having some fairness properties, expressed in most cases by groups of axioms. In an earlier work, we solved what we called the Inverse Problem for Semivalues, in which the main result was offering an explicit formula providing the set of all games with an a priori given Semivalue, associated with a given weight vector. However, in this set there is an infinite set of games for which the Semivalues are not coalitional rational, perhaps not efficient, so that these are not fair practical solutions of the above fundamental problem. Among the Semivalues, coalitional rational solutions for the Shapley Value and the Banzhaf Value have been given in two more recent works. In the present paper, based upon a general potential basis, relative to Semivalues, for a given game and a given Semivalue, we solve the connected problem: in the Inverse Set, find out a game with the same Semivalue, which is also coalitional rational. Several examples will illustrate the corresponding numerical technique.
基金The work was supported in part by the National Natural Science Foundation of China(61572370)and the National Natural Science Function of Qinghai Province(2019-ZJ-7065,2017-ZJ-959Q)+1 种基金the MOE(Ministry of Education in China)Project of Humanities and Social Sciences(17YJCZH203)and the Natural Science Foundation of Hubei Province in China(2016CFB652).
文摘Wireless sensor networks(WSNs)are the major contributors to big data acquisition.The authenticity and integrity of the data are two most important basic requirements for various services based on big data.Data aggregation is a promising method to decrease operation cost for resource-constrained WSNs.However,the process of data acquisitions in WSNs are in open environments,data aggregation is vulnerable to more special security attacks with hiding feature and subjective fraudulence,such as coalition attack.Aimed to provide data authenticity and integrity protection for WSNs,an efficient and secure identity-based aggregate signature scheme(EIAS)is proposed in this paper.Rigorous security proof shows that our proposed scheme can be secure against all kinds of attacks.The performance comparisons shows EIAS has clear advantages in term of computation cost and communication cost when compared with similar data aggregation scheme for WSNs.
文摘This article aims to explore the coalition of external actors and the strategies it deployed to influence the emergence of the National Nutrition Policy (NNP) in Lao People’s Democratic Republic (Lao PDR). The Advocacy Coalition Framework and the conceptual model of Effective Advocacy Strategies for Influencing Government Nutrition Policy were used to frame the data collection and their analysis. Sources of information were semi-structured interviews conducted with government and external actors, as well as all available documents on nutrition policy in Laos. The commitment of the government to achieve the Millennium Development Goals (MDGs) and to leave the Least Developed Country status created a favorable condition to support the emergence of the NNP in Laos. This context was a driving force for the building of an effective and convincing coalition of United Nations agencies able to accompany the government in redefining health priorities. Various strategies were used by the coalition to this end, including generating, disseminating, and using scientific evidence, assisting the government with a budget and technical expertise, providing decision-makers with opportunities to learn from other countries, and building relationships with the key actor. External actors can be a major force to support the emergence of a public policy in Laos, but this requires a window of opportunity like what the MDGs have been able to bring.
文摘The effective classification of urban domestic waste is the key to achieve a “waste-free city” and provides an essential guarantee for resource utilization. This article takes a coalitional game perspective to study the dilemmas in urban domestic waste separation from the cooperative interaction of residents, government, and enterprises. The study finds that urban domestic waste classification in China is currently facing many problems, focusing on: 1) insufficient consensus among residents, 2) shortage of input funds, 3) corporate profitability difficulties, 4) weak policy constraints, and 5) difficulties in integrating goals. In this regard, each participating body still needs to focus on collective interests, coalitional games, break the dilemma society, and promote the long-term management of urban domestic waste.
基金Supported by the National Natural Science Foundation of China (No. 60772062)the National High Technology Research and Development Program of China (No. 2009AA012241)Zhejiang Provincial Natural Science Foundation of China (No. Y1080935)
文摘Coalition game theory is introduced to investigate the performance,fairness and stability of decorrelating group multiuser detection receiver,not only from the perspective of individual nodes,but also various coalitions and the whole system as well. Firstly,to derive how the system scale with coalition size,a stochastic model with transferable payoffs (stochastic TU-model) is provided. Secondly,to find the most preferred coalition structures from the view point of individual nodes,a model with Non-Transferable payoffs (NTU-model) is presented. Theoretical analysis and simulation results suggest that stochasticaly the grand coalition is payoff maximizing for the system as a whole,while individual nodes with good-conditioned channels may prefer local "win-win coalitions".
文摘This paper proposes a multi-objective optimization design method based on the coalition cooperative game theory where the three design goals have been seen as three game players. By calculating the affecting factors and fuzzy clustering, the design variables are divided into different strategic spaces which belong to each player, then it constructs a payoff function based on the coalition mechanism. Each game player takes its own revenue function as a target and obtains the best strategy versus other players. The best strategies of all players consist of the strategy permutation of a round game and it obtains the final game solutions through multi-round games according to the convergence criterion. A multi-objective optimization example of the luff mechanism of compensative sheave block shows the effectiveness of the coalition cooperative game method.
