Regional Multi-Agent Cooperative Reinforcement Learning for City-Level Traffic Grid Signal Control

This article studies the effective traffic signal control problem of multiple intersections in a city-level traffic system.A novel regional multi-agent cooperative reinforcement learning algorithm called RegionSTLight is proposed to improve the traffic efficiency.Firstly a regional multi-agent Q-learning framework is proposed,which can equivalently decompose the global Q value of the traffic system into the local values of several regions Based on the framework and the idea of human-machine cooperation,a dynamic zoning method is designed to divide the traffic network into several strong-coupled regions according to realtime traffic flow densities.In order to achieve better cooperation inside each region,a lightweight spatio-temporal fusion feature extraction network is designed.The experiments in synthetic real-world and city-level scenarios show that the proposed RegionS TLight converges more quickly,is more stable,and obtains better asymptotic performance compared to state-of-theart models.

STABILITY ANALYSIS AND COOPERATIVE CONTROL OF DISTRIBUTED MULTI-AGENT SYSTEM WITH SAMPLED COMMUNICATION

The cooperative control and stability analysis problems for the multi-agent system with sampled com- munication are investigated. Distributed state feedback controllers are adopted for the cooperation of networked agents. A theorem in the form of linear matrix inequalities（LMI） is derived to analyze the system stability. An- other theorem in the form of optimization problem subject to LMI constraints is proposed to design the controller, and then the algorithm is presented. The simulation results verify the validity and the effectiveness of the pro- posed approach.

Multi-agent based modeling for order distribution cooperative negotiation in supply chains

With the new characteristics of global cooperation in supply chains being synthetically considered,a hybrid model to the cooperative negotiation process for the order distribution in supply chain is mainly studied.After reviewing and analyzing some main domestic and overseas processes in cooperative negotiation modeling in supply chain,some problems are subsequently pointed out.For example,the traditional simple multi-agent system(MAS)frameworks which have some limitations,are not suitable for solving modeling complex systems.To solve these problems,thinking with the aid of the multi-agent structure and complex system modeling,the manufacturing supply chain is taken as an example,and a time Petri net production model is adopted to decompose the materials.And then a cooperative negotiation model for the order distribution in supply chain is constructed based on combining multi-agent techniques with time Petri net modeling.The simulation results reveal that the above model helps solve the problems of cooperative negotiation in supply chains.

Immune multi-agent model using vaccine for cooperative air-defense system of systems for surface warship formation based on danger theory

Aiming at the problem on cooperative air-defense of surface warship formation, this paper maps the cooperative airdefense system of systems （SoS） for surface warship formation （CASoSSWF） to the biological immune system （BIS） according to the similarity of the defense mechanism and characteristics between the CASoSSWF and the BIS, and then designs the models of components and the architecture for a monitoring agent, a regulating agent, a killer agent, a pre-warning agent and a communicating agent by making use of the theories and methods of the artificial immune system, the multi-agent system （MAS）, the vaccine and the danger theory （DT）. Moreover a new immune multi-agent model using vaccine based on DT （IMMUVBDT） for the cooperative air-defense SoS is advanced. The immune response and immune mechanism of the CASoSSWF are analyzed. The model has a capability of memory, evolution, commendable dynamic environment adaptability and self-learning, and embodies adequately the cooperative air-defense mechanism for the CASoSSWF. Therefore it shows a novel idea for the CASoSSWF which can provide conception models for a surface warship formation operation simulation system.

Analysis on Governance Structure of Farmers' Specialized Cooperatives in Sichuan Province

In this paper, 87 farmers' specialized cooperatives in Sichuan Province were investigated firstly, and then their governance structure was analyzed from the aspects of authority, decision-making mechanism, supervision mechanism and executive body based on the corporative governance structure in narrowest sense, finally corresponding policy suggestions for improving the governance structure of farmers' specialized cooperatives in Sichuan Province were put forward. It is concluded that human control is serious within farmers' specialized cooperatives; inner supervision is effective; council has great effects on governance structure; the cooperatives give limited incentive to managers.

Evaluation of Governance Structure of Farmers' Specialized Cooperatives

In this paper,from the authority,decision-making body,supervisory body and implementing body,we analyze the governance structure evaluation indicators of farmers' specialized cooperatives,using objective and subjective combination weighting approach.And we draw the following conclusions:the rate of return of profits for membership,frequency of council convened and the number of members of board of supervisors have a great impact on the governance structure of cooperatives; the insider control phenomenon in cooperatives is more serious and the level of incentives is limited.Finally we put forward corresponding policy recommendations.

Cooperative Multi-Agent Reinforcement Learning with Constraint-Reduced DCOP

Cooperative multi-agent reinforcement learning（ MARL） is an important topic in the field of artificial intelligence,in which distributed constraint optimization（ DCOP） algorithms have been widely used to coordinate the actions of multiple agents. However,dense communication among agents affects the practicability of DCOP algorithms. In this paper,we propose a novel DCOP algorithm dealing with the previous DCOP algorithms＇ communication problem by reducing constraints.The contributions of this paper are primarily threefold：（1） It is proved that removing constraints can effectively reduce the communication burden of DCOP algorithms.（2） An criterion is provided to identify insignificant constraints whose elimination doesn＇t have a great impact on the performance of the whole system.（3） A constraint-reduced DCOP algorithm is proposed by adopting a variant of spectral clustering algorithm to detect and eliminate the insignificant constraints. Our algorithm reduces the communication burdern of the benchmark DCOP algorithm while keeping its overall performance unaffected. The performance of constraint-reduced DCOP algorithm is evaluated on four configurations of cooperative sensor networks. The effectiveness of communication reduction is also verified by comparisons between the constraint-reduced DCOP and the benchmark DCOP.

A multiple case study on the cooperative governance model of treating environmental deterioration

Managing the acute problem of environmental deterioration due to the industrialization of modern China is necessary.Using three cases from a theoretical perspective of procedure-structure,this study explored how environmental deterioration is governed in three circumstances.Research findings revealed that the pattern to treat the worsening environment is associated closely with individual-specific circumstances,and thus three patterns of governance are developed:bureaucratic,contractual,and network-based cooperatives.The cases and the results are described the advantages and disadvantages of these models are discussed.Suggestions on how to more effectively manage these urgent problems are as follows:first,the governance must fit the circumstance,namely,its physical,biological,and social attributes;second,the regulations and conditions regarding the government-dominated pattern are came up from the perspective of mega-governance,mutual trust contracts,benefit sharing,and social capital;and third,the government should abandon the pattern based on the single subject and adopt a plural subject governance model to integrate all the relevant forces to facilitate the attainment of benign environmental governance.

Ocean ecological compensation mechanism in Liaoning area from the angle of ecological civilization——based on analysis of Dalian's cooperative governance

How to speed up the establishment of marine ecological compensation and the ecological compensation system is inevitable choice from ocean country to marine power.Therefore,construction of marine ecological compensation mechanism is the effective means which incentives marine ecosystem services supply,improves the quality of marine ecology,and promotes the ocean to the green transformation.Combined with Dalian maritime case study in Liaoning Province,three dimensional interactive marine ecological compensation mechanism,the compensation frame work,path and operation mechanism were built up.Suggestions on accelerating the legislation process;perfecting the management system;establishing proper system for both incentive constraint and reward-punishment;increasing the source of Ocean Ecological Compensation funds;accelerating On the study of the Ocean Ecological Compensation Mechanism in Liaoning Area from the Angl-e of Ecological Civilization Based on Case Analysis of Dalian.

Cooperative multi-target hunting by unmanned surface vehicles based on multi-agent reinforcement learning

To solve the problem of multi-target hunting by an unmanned surface vehicle(USV)fleet,a hunting algorithm based on multi-agent reinforcement learning is proposed.Firstly,the hunting environment and kinematic model without boundary constraints are built,and the criteria for successful target capture are given.Then,the cooperative hunting problem of a USV fleet is modeled as a decentralized partially observable Markov decision process(Dec-POMDP),and a distributed partially observable multitarget hunting Proximal Policy Optimization(DPOMH-PPO)algorithm applicable to USVs is proposed.In addition,an observation model,a reward function and the action space applicable to multi-target hunting tasks are designed.To deal with the dynamic change of observational feature dimension input by partially observable systems,a feature embedding block is proposed.By combining the two feature compression methods of column-wise max pooling(CMP)and column-wise average-pooling(CAP),observational feature encoding is established.Finally,the centralized training and decentralized execution framework is adopted to complete the training of hunting strategy.Each USV in the fleet shares the same policy and perform actions independently.Simulation experiments have verified the effectiveness of the DPOMH-PPO algorithm in the test scenarios with different numbers of USVs.Moreover,the advantages of the proposed model are comprehensively analyzed from the aspects of algorithm performance,migration effect in task scenarios and self-organization capability after being damaged,the potential deployment and application of DPOMH-PPO in the real environment is verified.

Cooperative output tracking of multi-agent systems under finite time

The cooperative output tracking problem of multi-agent systems in finite time is considered.In order to enable the agents to quickly track and converge to external system within a finite time,a novel distributed output feedback control strategy based on the finite-time state observer is designed.This distributed finite-time observer can not only solve cooperative output tracking problems when the agents can not get external system signal,but also make the systems have a faster convergence and a good robustness.The stability of the system in finite time is proved based on Lyapunov function.Numerical simulations results have been provided to demonstrate the effectiveness of the proposed protocol.

The Impact of Corporate Governance on Banks Performance and Loan Quality： Evidence From Italian Cooperative Banks

In the banking system, a context characterized by growing instability and by the speed of evolution of business dynamics, the system of corporate governance plays a key role, both for large banks and for banks with a smaller size. The paper aims to investigate the influence of corporate governance of the banks that operate in the cooperative credit system on performance and quality of loans, over the years 2010-2011-2012. In order to achieve this aim, the following research hypotheses have been formulated： There is a statistically significant relationship between the size and the structure of the board and banks Performance operating in the cooperative credit system; there is a statistically significant relationship between the size and structure of the board and the credit quality of banks operating in the cooperative credit system. The analysis is conducted on a sample composed of 48 Italian banks, divided into 24 cooperative banks and 24 popular banks. The sample is made up of banks from all over Italy and very different in terms of dimension. For the sample construction, data relative to governance of banks were collected from the balance sheets in the period from June to August 2014. Financial-economic data were collected from the Financial Statements and from Bankscope database. The methodology of analysis is based on multivariate OLS （ordinary least squares） regression models. The main results refer to the presence of significant relationships between board dimension and the quality of loans and among the number of committees and performance and the quality of loans. The presence of a significant and negative relationship between the board dimension and the ratio of impaired loans to gross loans indicates the possibility that enlargement of board dimension allows a better quality of loans. The presence of a positive relationship between the number of committees and the ratio of impaired loans to gross loans signals the possibility that a greater number of committees can produce a worse quality of loans. The presence of a negative relationship between the number of committees and bank performance suggests to limit and manage the complexity of governance in banks operating in the cooperative credit system.

Cooperative Governance Path of Financial Technology Supervision

Financial technology changes the logic of financial interpretation through the use of digital and digital centric technologies,commercialization,big data analysis,machine learning and artificial intelligence.From financial institutions that use technology to provide financial services to technology companies that directly provide financial services,fintech companies play an important role in realizing financial brokerage and financial democratization and improving the availability and efficiency of financial services.Based on this,this paper focuses on the plight and path of cooperative governance of financial technology supervision,for the reference of relevant personnel.

The Governance Structure and Performance of Agricultural Cooperative Economic Organizations in the Three Gorges Reservoir Area

On the basis of questionnaire survey results of 100 agricultural cooperative economic organizations in Three Gorges Reservoir Area,this article uses the structural equation modeling as analysis tool to conduct empirical analysis of the governance structure and performance of agricultural cooperative economic organizations. The results show that the ownership structure has the most critical impact on the performance of cooperative organizations,and the impact coefficient is as high as 0.92; the importance of oversight mechanism is basically equivalent to that of the council structure for cooperative organizations,and the impact coefficient is 0.87 and 0.86,respectively,second only to the ownership structure; the size of members' quit capacity also has a very important impact on the performance of organizations,with impact coefficient of 0.74. Thus,the problems influencing the performance of organizations,in the current governance structure of agricultural cooperative economic organizations in Three Gorges Reservoir Area,are pointed out. Finally,corresponding countermeasures are put forward.

Robust adaptive leaderless consensus of unknown non-minimum phase linear multi-agent systems subject to disturbances and/or unmodeled dynamics

This article investigates the problem of robust adaptive leaderless consensus for heterogeneous uncertain nonminimumphase linear multi-agent systems over directed communication graphs. Each agent is assumed tobe of unknown nominal dynamics and also subject to external disturbances and/or unmodeled dynamics. Anovel distributed robust adaptive control strategy is proposed. It is shown that the robust adaptive leaderlessconsensus problem is solved with the proposed control strategy under some sufficient conditions. Two examplesare provided to demonstrate the efficacy of the proposed control strategy.

Trading platform for cooperation and sharing based on blockchain within multi-agent energy internet

With the release of the electricity sales side,large-scale small-capacity distributed power generation units are connected to the distribution side,forming multi-type market entities such as microgrids,integrated energy systems,and virtual power plants.With the large-scale integration of distributed energy,the energy market under the energy internet is different from a traditional transmission grid.It is currently developing in the direction of diversified entities and commodities,a flat structure,and a flexible and competitive multi-agent market mechanism.In this context,this study analyzes the value of combining blockchain and the electricity market presents the design of a blockchain trading framework for multi-agent cooperation and sharing of the energy internet.The nodes in market transactions are modeled through power system modeling in the physical layer and the transaction consensus strategy in the cyber layer;moreover,the nodes are verified in a modified IEEE 13 testing feeder of a distribution network.A transaction example is demonstrated using the multi-agent cooperation and sharing transaction platform based on the Ethereum private blockchain.

Evolution Characteristics of Government-Industry-University Cooperative Innovation Network of Electronic Information Industry in Liaoning Province, China

It is important to optimize the cooperative innovation network for the improvement of economic competence and innovative power. Based on the patent information services platform, we obtain invention patent data for the electronic information industry in Liaoning from 1985 to 2015. This paper analyzes the cooperative innovation network structure, its spatiotemporal evolution and the triple helix relationship of government-industry-university(GIU) by using the social network analysis method and the triple helix theory as well as UCINet, ArcGIS and NetDraw. The empirical results show that: 1) the number of the subjects of the electronic information industry GIU cooperative innovation network in Liaoning demonstrates a gradual increase from 1985 to 2015, with the same trend in concentration. In terms of its subject and its centrality, the universities have a higher position, and the industries have a lower position,while the status of the government is still unclear. 2) The cooperative innovation network presents a core-periphery structure, and the polarization effect of innovation subjects tends to be obvious. There is certain distance-decay regularity in the cooperative innovation network, and a strong geographical proximity to cooperative innovation. 3) The compactness shows a downward trend as a whole. In terms of the extent of participation, the industries are better than the government but worse than the universities. This means that the cooperative innovation network of GIU in the electronic information industry in Liaoning is in the initial stage of formation.

Resource Dependence Relationship between Grass-roots Government and Farmers' Specialized Cooperatives

Based on theories of resource dependence and government functions,this paper analyzed benefit interaction between grass-roots government and farmers' specialized cooperatives in Hubei Province. It was intended to make clear the path dependence relationship between grassroots government and farmers' specialized cooperatives and find out problems in between. Besides,from the perspective of grass-roots government and farmers' specialized cooperatives,it came up with corresponding recommendations,to promote constant improvement of farmers' specialized cooperatives,and help them to become key forces of leading farmers to get rich,pull agricultural economic development,and promote new socialist countryside construction.

Research on Cognitive Cooperation in Multi-Agent Systems

The accomplishment of a complex problem usually involves cooperation between participators with different knowledge background concerned. This paper identifies interdependency between different sub problems （through problem decomposition） as the major factor that influences cooperative relations in multi-Agent systems, based on which we propose an efficient means to measure cooperation coefficient （degree） between different Agents. Then cognitive cooperation between Agents is analyzed which aims at collecting the wisdom of the cognitive community for a systematic solution to the overall problem.

Rough set based multi-agent system cooperation for industrial supervisory interface system

In this paper, rough set theory is introduced into the interface multi-agent system （MAS） for industrial supervisory system. Taking advantages of rough set in data mining, a cooperation model for MAS is built. Rules for avoiding cooperation conflict are deduced. An optimization algorithm is used to enhance security and real time attributes of the system. An application based on the proposed algorithm and rules are given.