Metabolism of Social System--论文代写范文精选

然后系统理论用于描述个体，在社会系统交互沟通,在这种情况下的囚徒困境,个体之间的相互作用视为数学模型，只通过看这些交互解释的行动选择合作还是不合作。下面的essay代写范文进行详述。

Abstract 
Random Boolean Network has been used to find out regulation patterns of genes in organism. This approach is very interesting to use in a game such as N-Person Prisoner’s Dilemma. Here we assume that agent’s action is influenced by input in the form of choices of cooperate or defect she accepted from other agent or group of agents in the system. Number of cooperators, pay-off value received by each agent, and average value of the group, are observed in every state, from initial state chosen until it reaches its state-cycle attractor. In simulation performed here, we gain information that a system with large number agents based on action on input K equals to two, will reach equilibrium and stable condition over strategies taken out by its agents faster than higher input, that is K equals to three. Equilibrium reached in longer interval, yet it is stable over strategies carried out by agents. 
Keywords: evolutionary game theory, N-Person Prisoner’s Dilemma, cooperation, social stability, Random Boolean Network, Transition Rule

Introduction 
One of interesting aspects in understanding complex social system is about how simple interactions occur among elements constitutes it: human individual interactions emerge as a complex macro phenomena (Paczuski, 2000). In Sociology, this concept later known as human agency (Fuch, 2002; Situngkir, 2003a) or structure of macro-micro linkage in social system (Sawyer, 2001; Situngkir, 2003b). Macro-micro linkage of the social system stated that macro phenomenon in the form of social structure (norms, social behaviors, organization, etc) are emergent phenomena coming out from interactions occurring in individual/micro level (Situngkir, 2003a; Frumkin, et al). The idea is similar to some sociologists who support structuration theory such as Giddens (1984) and Minger who also see that social structure is a macro phenomena resulted from an individual level interactions process (Fuch, 2002). However, the problem is getting more complicated as we try to see the intrapersonal elementary interactions form that constitute social system. Niklas Luhman (1990) suggests that communication is the elementary form of intrapersonal interaction in social system. 

He then used autopoietic system theory to describe how individuals interact in social system through communication (Viskovatoff, 1999). In Game Theoretic Analysis, in this case Prisoner’s Dilemma, interactions among individuals viewed as mathematic model where those interactions interpreted only through watching its action of choose to cooperate or not to cooperate (defect). Simple interaction occurred in Prisoner’s Dilemma is believed to be a form of more elementary interaction than another more complicated intrapersonal interactions. In analysis using Prisoner’s Dilemma, every individual action is not based on certain social structure; name it social norms, social behavior/habit or even social sanction (Robinson et al, 2000), henceforth we can virtually see whether certain social structure such as cooperation property is likely to emerge as a collective character from selfish agents that play. Human agency concept stating that social structure may emerge through interactions among their agents is highly appropriate in case of Prisoner’s Dilemma game. This game also has been known to be utilized in understanding complex phenomena occur in social and field of economy (Cho, 2000).

N-person Iterated Prisoner’s Dilemma
By noticing those pay-offs, in one round play, whatever action taken by other agent, each agent will tend to choose cooperate (D) – Nash equilibrium from this single play is easily understand that agent will tend to cooperate. However, if the game has a chance to play once more, then each agent will evaluate payoff gained and chosen in the previous round. Such game is called iterated Prisoner’s Dilemma. This IPD model is then developed into a game involving higher number of agents, be it for 3 persons (Akiyama, et al, 1995) or for N-person (Axelrod, 1997; Akimov, 1994; and Lindgren, 1998). This kind of game has been widely used to explain the emergence of cooperation in group of agents where every agent in the group is selfish and interacting playing Prisoner’s Dilemma game.

Model Inspiration
From the above explanation, we perform a set of experiment to see how IPD applied in RBN model. We will see and analyze Alexander model and compare it to the actual RBN model with some needed modifications on implementation to the RBN model so that it will really describe individual interactions as represented in IPD. In the ring model by Alexander (2002), the translation of N-person Prisoner’s Dilemma was performed in Random Boolean Network fashion. He assumed agents play Prisoner’s Dilemma in a network with ring formation, where each agent updates its state by rule: imitating winner neighbor state in order each agent calculates how much pay-off gained by every of his neighbors. Every agent assumed to know other agent state that is other neighbor’s neighbor. At that condition, agents playing in ring model described as a node that has input from its 2 closest neighbors node and the node itself (Figure 5). In this ring PD game, every agent supposed to link with 2 neighbors and play 2- person PD game. Pay-off obtained by agent only encountered only from neighbor agents, thus the social dilemma can not be described in such fashion. 

The purpose of Alexander (2002) in the ring model was to describe the basin attractor in the PD game. However, it was not explained what the connection between attractor valley he described with existing social phenomena. That kind of model was aimed to seek attractor state in every agent condition but this is rather difficult to do when the numbers of agents used are too large. If we look carefully to the model, we will find some weakness since the model seems to be unrealistic, regarding the fix and pre-defined construction. However, the basic idea of this model is sufficiently describing how IPD viewed in RBN fashion. Different with Alexander (2002), Paczuski, et. al. (2000) described agents as well as in Random Boolean Network – with a goal to describe the economic model, i.e.: minority game, in RBN fashion. They viewed that every agent will ground her action on limited information he received from other agent or other group of agent. This condition was described in NK Random Boolean Network, where N represents the number of agent and K represents the number of input of agents or other group of agent. Agent’s strategy is made based on input she has, that picked from the possible Boolean Function. At every round every agent will gain reward and punishment. After certain interval, the looser, that is agent lies in majority group, will change her strategy randomly.

Discussion 
In the first simulation, for agent with pre-defined two neighbors and subjected rule transition of imitating the state of the winning neighbor, we can see that the system will tend to earn equilibrium point in the defect regime. This condition is surely will be not beneficial since the system is heading destruction. Furthermore, this model does not use RBN as a genetic model; yet, this model has potential to be utilized to finding attractor and basin attractor of the modeled social system. N=20 N=20 N=30 N=100 Figure 10 Pay-off for each agent for different number of agents Figure 11 Experimental result with N=20 and K=3. The result tends to cyclic at equilibrium as the rise of homogeneity parameter 12 In the second simulation, we have described the behavior of the system whose bounded agent defect and cooperate with her neighbor agent. Every agent, based on her transition rule– choose whether to cooperate/defect (C/D) with her neighbor. At certain interval of time (state cycle length is assumed), every agent evaluate the result she gained in the form of average pay-off at the interval of time and change her strategy by using the strategy of her neighbor with highest average pay-off.

Conclusion 
We have shown how to perform iteration over NIPD in RBN fashion. In several experiments performed we show the potential of RBN model to be utilized as alternative genetic model – as claimed by Kauffman in (Paczuski, et.al., 2000). The smaller amount of agents whose high homogeneity will perform more stable system, a conclusion that showing that in a homogenous society conflict probability is seldom– agent tends to cooperate and system direction can be “predicted” easily. In return, social system becomes mode dynamic with the rise amount of agents. However, collective system will tend to reach its equilibrium (with certain pay-off) although it has to through a long process (large attractor length). This is the social system behavior performing the robustness in society and the fact of the self-organizing system. The weakness of this paper is known comes from the lack computational facilities good enough by means that further works will highly depend on better facilities to gain better conclusion and theoretical exploration. In short, we can say that RBN is an alternative to construct evolutionary model of social system with complex characters like self-organizing, adaptive, and evolve dynamically. These characters are more or less theoretically explained by RBN for organism metabolism widely – this paper is a mean of hope the same thing to view social system metabolism to be observed further.

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。(论文代写)
更多essay代写范文欢迎访问我们主页 www.51due.com 当然有essay代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X(论文代写)