Sub-strategy updating evolution in minority game

In this paper, we propose and study a new evolution model of minority game. Any strategy in minority game can be regarded as composed of sub-strategies corresponding to different histories. Based on the evolution model proposed by Li－Riolo－Savit, in which those agents that perform poorly may update their strategies randomly. This paper presents a new evolution model in which poor agents update their strategies by changing only a part of sub-strategy sets with low success rate. Simulation result shows that the new model with sub-strategy-set updating evolution mechanism may approach its steady state more quickly than the Li－Riolo－Savit model. In the steady state of the new model, stronger adaptive cooperation among agents will appear, implying that the social resource can be allocated more rationally and utilized more effectively compared with the Li－Riolo－Savit model.     

Network evolution based on minority game with herding behavior

The minority game (MG) is used as a source of information to design complex networks where the nodes represent the playing agents. Differently from classical MG consisting of independent agents, the current model rules that connections between nodes are dynamically inserted or removed from the network according to the most recent game outputs. This way, preferential attachment based on the concept of social distance is controlled by the agents wealth. The time evolution of the network topology, quantitatively measured by usual parameters, is characterized by a transient phase followed by a steady state, where the network properties remain constant. Changes in the local landscapes around individual nodes depend on the parameters used to control network links. If agents are allowed to access the strategies of their network neighbors, a feedback effect on the network structure and game outputs is observed. Such effect, known as herding behavior, considerably changes the dependence of volatility σ on memory size: it is shown that the absolute value of σ as well as the corresponding value of memory size depend both on the network topology and on the way along which the agents make their playing decisions in each game round.     

Promotion of cooperation induced by the interplay between structure and game dynamics

We consider the coupled dynamics of the adaption of network structure and the evolution of strategies played by individuals occupying the network vertices. We propose a computational model in which each agent plays a n-round Prisoner's Dilemma game with its immediate neighbors, after that, based upon self-interest, partial individuals may punish their defective neighbors by dismissing the social tie to the one who defects the most times, meanwhile seek for a new partner at random from the neighbors of the punished agent. It is found that the promotion of cooperation is attributed to the entangled evolution of individual strategy and network structure. Moreover, we show that the emerging social networks exhibit high heterogeneity and disassortative mixing pattern. For a given average connectivity of the population and the number of rounds, there is a critical value for the fraction of individuals adapting their social interactions, above which cooperators wipe out defectors. Besides, the effects of the average degree, the number of rounds, and the intensity of selection are investigated by extensive numerical simulations. Our results to some extent reflect the underlying mechanism promoting cooperation.     

Evolutionary Prisoner's Dilemma Game Based on Pursuing Higher Average Payoff

We investigate the prisoner＇s dilemma game based on a new rule： players will change their current strategies to opposite strategies with some probability if their neighbours＇ average payoffs are higher than theirs. Compared with the cases on regular lattices （RL） and Newman-Watts small world network （NW）, cooperation can be best enhanced on the scale-free Barabasi-Albert network （BA）. It is found that cooperators are dispersive on RL network, which is different from previously reported results that cooperators will form large clusters to resist the invasion of defectors. Cooperative behaviours on the BA network are discussed in detail. It is found that large-degree individuals have lower cooperation level and gain higher average payoffs than that of small-degree individuals. In addition, we find that small-degree individuals more frequently change strategies than do large- degree individuals.     

Study on system dynamics of evolutionary mix-game models

Mix-game model is ameliorated from an agent-based MG model, which is used to simulate the real financial market. Different from MG, there are two groups of agents in Mix-game: Group 1 plays a majority game and Group 2 plays a minority game. These two groups of agents have different bounded abilities to deal with historical information and to count their own performance. In this paper, we modify Mix-game model by assigning the evolution abilities to agents: if the winning rates of agents are smaller than a threshold, they will copy the best strategies the other agent has; and agents will repeat such evolution at certain time intervals. Through simulations this paper finds: (1) the average winning rates of agents in Group 1 and the mean volatilities increase with the increases of the thresholds of Group 1; (2) the average winning rates of both groups decrease but the mean volatilities of system increase with the increase of the thresholds of Group 2; (3) the thresholds of Group 2 have greater impact on system dynamics than the thresholds of Group 1; (4) the characteristics of system dynamics under different time intervals of strategy change are similar to each other qualitatively, but they are different quantitatively; (5) As the time interval of strategy change increases from 1 to 20, the system behaves more and more stable and the performances of agents in both groups become better also.     

Spatial Games Based on Pursuing the Highest Average Payoff

We propose a strategy updating mechanism based on pursuing the highest average payoff to investigate the prisoner＇s dilemma game and the snowdrift game. We apply the new rule to investigate cooperative behaviours on regular, small-world, scale-free networks, and find spatial structure can maintain cooperation for the prisoner＇s dilemma game. fn the snowdrift game, spatial structure can inhibit or promote cooperative behaviour which depends on payoff parameter. We further study cooperative behaviour on scale-free network in detail. Interestingly, non-monotonous behaviours observed on scale-free network with middle-degree individuals have the lowest cooperation level. We also find that large-degree individuals change their strategies more frequently for both games.     

Intelligent minority game with genetic crossover strategies

We have developed a novel game theoretical model of N interacting agents playing a minority game such that they change their strategies intelligently or adaptively depending on their temporal performances. The strategy changes are done by generating new strategies through one-point genetic crossover mechanism. The performances of agents are found to change dramatically (from losing to winning or otherwise) and the game moves rapidly to an efficient state, in which fluctuations in the number of agents performing a particular action, characterized by the variance , reaches a low value.Received: 2 June 2003, Published online: 11 August 2003PACS: 87.23.Ge Dynamics of social systems - 02.50.Le Decision theory and game theory - 87.23.Kg Dynamics of evolution     

Cooperation in networked populations of selfish adaptive agents: Sensitivity to network structure

This paper investigates the adaptation of cooperating strategies in an iterated prisoner's dilemma (IPD) game with individually learning agents, subject to the structure of the interaction network. In particular, we study how cooperation or defection comes to dominate the population on Watts–Strogatz networks, under varying average path lengths. Our results are in good agreement with previous works on discrete choice dynamics on networks, but are in stark contrast with results from the evolution of cooperation literature. We argue that the latter is because the different adaptation method used (i.e., adaptive learning instead of ‘evolutionary’ strategy switching).     

Stability of strategies in payoff-driven evolutionary games on networks

We consider a network of coupled agents playing the Prisoner's Dilemma game, in which players are allowed to pick a strategy in the interval [0, 1], with 0 corresponding to defection, 1 to cooperation, and intermediate values representing mixed strategies in which each player may act as a cooperator or a defector over a large number of interactions with a certain probability. Our model is payoff-driven, i.e., we assume that the level of accumulated payoff at each node is a relevant parameter in the selection of strategies. Also, we consider that each player chooses his∕her strategy in a context of limited information. We present a deterministic nonlinear model for the evolution of strategies. We show that the final strategies depend on the network structure and on the choice of the parameters of the game. We find that polarized strategies (pure cooperator∕defector states) typically emerge when (i) the network connections are sparse, (ii) the network degree distribution is heterogeneous, (iii) the network is assortative, and surprisingly, (iv) the benefit of cooperation is high.     

Stratified economic exchange on networks

We investigate a model of stratified economic interactions between agents when the notion of spatial location is introduced. The agents are placed on a network with near-neighbor connections. Interactions between neighbors can occur only if the difference in their wealth is less than a threshold value that defines the width of the economic classes. By employing concepts from spatiotemporal dynamical systems, three types of patterns can be identified in the system as parameters are varied: laminar, intermittent and turbulent states. The transition from the laminar state to the turbulent state is characterized by the activity of the system, a quantity that measures the average exchange of wealth over long times. The degree of inequality in the wealth distribution for different parameter values is characterized by the Gini coefficient. High levels of activity are associated to low values of the Gini coefficient. It is found that the topological properties of the network have little effect on the activity of the system, but the Gini coefficient increases when the clustering coefficient of the network is increased.     

Effects of decision-making on the transport costs across complex networks

We analyse the effects of agents’ decisions on the creation of congestion on a centralised network with ring-and-hub topology. We show that there are two classes of agents each displaying a distinct set of behaviours. The dynamics of the system are driven by an interplay between the formation of, and transition between, unique stable states that arise as the network is varied. We show how the flow of objects across the network can be understood in terms of the ordering and allocation of strategies. Our results show that the existence of congestion in a network is a dynamic process that is as much dependent on the agents’ decisions as it is on the structure of the network itself.     

Dynamical phase transitions in Hegselmann-Krause model of opinion dynamics and consensus

The dynamics of the model of agents with limited confidence introduced by Hegselmann and Krause exhibits multiple well-separated regimes characterised by the number of distinct clusters in the stationary state. We present indications that there are genuine dynamical phase transitions between these regimes. The main indicator is the divergence of the average evolution time required to reach the stationary state. The slowdown close to the transition is connected with the emergence of the groups of mediator agents which are very small but have decisive role in the process of social convergence. More detailed study shows that the histogram of the evolution times is composed of several peaks. These peaks are unambiguously interpreted as corresponding to mediator groups consisting of one, two, three etc. agents. Detailed study reveals that each transition possesses also an internal fine structure.     

Coevolution of agents and networks: Opinion spreading and community disconnection

We study a stochastic model for the coevolution of a process of opinion formation in a population of agents and the network which underlies their interaction. Interaction links can break when agents fail to reach an opinion agreement. The structure of the network and the distribution of opinions over the population evolve towards a state where the population is divided into disconnected communities whose agents share the same opinion. The statistical properties of this final state vary considerably as the model parameters are changed. Community sizes and their internal connectivity are the quantities used to characterize such variations.     

Coevolving agent strategies and network topology for the public goods games

Much of human cooperation remains an evolutionary riddle. Coevolutionary public goods games in structured populations are studied where players can change from an unproductive public goods game to a productive one, by evaluating the productivity of the public goods games. In our model, each individual participates in games organized by its neighborhood plus by itself. Coevolution here refers to an evolutionary process entailing both deletion of existing links and addition of new links between agents that accompanies the evolution of their strategies. Furthermore, we investigate the effects of time scale separation of strategy and structure on cooperation level. This study presents the following: Foremost, we observe that high cooperation levels in public goods interactions are attained by the entangled coevolution of strategy and structure. Presented results also confirm that the resulting networks show many features of real systems, such as cooperative behavior and hierarchical clustering. The heterogeneity of the interaction network is held responsible for the observed promotion of cooperation. We hope our work may offer an explanation for the origin of large-scale cooperative behavior among unrelated individuals.     

Synchronization and structure in an adaptive oscillator network

We analyze the interplay of synchronization and structure evolution in an evolving network of phase oscillators. An initially random network is adaptively rewired according to the dynamical coherence of the oscillators, in order to enhance their mutual synchronization. We show that the evolving network reaches a small-world structure. Its clustering coefficient attains a maximum for an intermediate intensity of the coupling between oscillators, where a rich diversity of synchronized oscillator groups is observed. In the stationary state, these synchronized groups are directly associated with network clusters.     

基于无标度网络拓扑结构变化的舆论演化模型

基于BA无标度网络,构建了舆论和网络结构相互影响的自适应舆论演化模型.舆论的演化不仅受制于网络拓扑结构,而且也导致拓扑结构的变化.研究表明,演化达到稳态后,网络结构不再是初始的无标度网络而呈现泊松分布,而且随着时间的推移,系统中的舆论演化表现出很强的趋同效应,原来初始状态的几十个舆论,在长时间的演化后,大部分舆论灭亡,只有少数的舆论存留,且发展壮大.这种趋向与社会上的舆论、意见、信仰的演化大体上是符合的.     

Local Minority Game with Evolutionary Strategies

We study a model of local minority game in the random Kauffman network with evolutionary strategies and propose three methods to update the strategy of poor agents, with lower points in a given generation： namely to update either the Boolean function of their strategies randomly, or their local information of randomly adjacent m agents, or the number m of randomly chosen adjacent agents. The results of extended numerical simulations show that the behaviour of strategies in the three methods may enhance significantly the entire coordination of agents in the system. It is also found that a poor agent tends to use both small m strategies and correlated strategies, and the strategies of agents will finally self-organize into a steady-state distribution for a long time playing of the game.     

Frequency clustering of coupled phase oscillators on small-world networks

We analyze the phenomenon of frequency clustering in a system of coupled phase oscillators. The oscillators, which in the absence of coupling have uniformly distributed natural frequencies, are coupled through a small-world network, built according to the Watts-Strogatz model. We study the time evolution and determine variations in the transient times depending on the disorder of the network and on the coupling strength. We investigate the effects of fluctuations in the average frequencies, and discuss the definition of the threshold for synchronization. We characterize the structure of clusters and the distribution of cluster sizes in the synchronization transition, and define suitable order parameters to describe the aggregation of the oscillators as the network disorder and the coupling strength change. The non-monotonic behavior observed in some order parameters is related to fluctuations in the mean frequencies.     

Outward-inward information flux in an opinion formation model on different topologies

A simple model of opinion formation dynamics in which binary-state agents make up their opinions due to the influence of agents in a local neighborhood is studied using different network topologies. Each agent uses two different strategies, the Sznajd rule with a probability q and the Galam majority rule (without inertia) otherwise; being q a parameter of the system. Initially, the binary-state agents may have opinions (at random) against or in favor about a certain topic. The time evolution of the system is studied using different network topologies, starting from different initial opinion densities. A transition from consensus in one opinion to the other is found at the same percentage of initial distribution no matter which type of network is used or which opinion formation rule is used.     

Quantum strategies win in a defector-dominated population

Quantum strategies are introduced into evolutionary games. The agents using quantum strategies are regarded as invaders, whose fraction generally is 1% of a population, in contrast to the 50% of the population that are defectors. In this paper, the evolution of strategies on networks is investigated in a defector-dominated population, when three networks (square lattice, Newman–Watts small-world network, and scale-free network) are constructed and three games (Prisoners’ Dilemma, Snowdrift, and Stag-Hunt) are employed. As far as these three games are concerned, the results show that quantum strategies can always invade the population successfully. Comparing the three networks, we find that the square lattice is most easily invaded by agents that adopt quantum strategies. However, a scale-free network can be invaded by agents adopting quantum strategies only if a hub is occupied by an agent with a quantum strategy or if the fraction of agents with quantum strategies in the population is significant.