Model of Co-evolving Social Agents--论文代写范文精选

每个代理不知道任何表达式的意义或实用程序，它只能评估整个表达式的效用。这是唯一的方式影响模型，这意味着任何特定的表达没有先天的意义。这尤其对于表达式确定代理的沟通。下面的essay代写范文进行详述。

The model is based upon Brian Arthur’s ‘El Farol Bar’ model [2], but extended in several respects, principally by introducing learning and communication. There is a fixed population of agents (in this case 10). Each week each agent has to decide whether or not to go to El Farol’s Bar on thursday night. Generally, it is advantageous for an agent to go unless it is too crowded, which it is if 67% or more of all the agents go (in this case 7 or more). This advantage is expressed as a utility, but this only impacts on the model in the agents evaluations of their constructs. Before making their decision agents have a chance to communicate with each other. This model can be seen as an extension of the work in [1], which investigates a three player game.

The environment
There are two alternative schemes for representing the utility gained by agents, which I have called: friendly and crowd-avoiding. In the crowd-avoiding scheme each agent gets the most utility for going when less than 7 of the other agents go (0.7), they get a fixed utility (0.5) if they do not go and the lowest utility for going when it is crowded (0.4). In this way there is no fixed reward for any particular action because the utility gained from going depends on whether too many other agents also go. In this way there is no fixed goal for the agent’s learning, but it is relative to the other agent’s behaviour (which will, of course, change over time). Under this scheme it is in each agent’s interest to discoordinate their action with the others (or, at least, a majority of the others).

The friendly scheme is similar to the crowd-avoiding scheme, there is a basic utility of 0.5 for going if it is not crowded, and 0.2 if it is but if they go to the bar each agents gets a bonus (0.2) for each ‘friend’ that also goes. If they stay at home they are guarenteed a utility of 0.65, so it is worth going if you go when it is not crowded with at least one other friend or if it is crowded with 3 or more friends. Who is a friend of whom is decided randomly at the beginning and remains fixed thereafter. Friendship is commutative, that is if A is a friend of B then B is a friend of A. An example of such a network is illustrated in figure 2. The number of friendships and nodes is constant accross runs but the detailed structure dif feres. In this scheme it is in the interest of agents to go when their other friends and only their friends are going. Under this scheme it is in each agent’ s interest to coordinate its actions with its designated friends but to discoordinate its action with the other agents.

The agents Each agent has a population of (pairs of) expressions that represent possible behaviours in terms of what to say and what to do (its constructs). This population is fixed in size but not in content. These expressions are taken from a strongly typed formal language which is specified by the programmer, but the expression can be of any structure and depth. Each agent does not ‘know’ the meaning or utility of any expression, communication or action – it can only evaluate each whole expression as to the utility each expression would have resulted in if it had used it in the past to determine whether it would go to the bar or not and the other’s behaviours had remained the same.

This is the only way in which the utilities affect the course of the model. Each week each agent takes the best such pair of expressions (in terms of its present evaluation against the recent past history) and uses them to determine its communication and action. 2 7 8 6 1 5 10 4 3 9 - page 9 - This means that any particular expression does not have an a priori meaning for that agent – any such meaning has to be learned. This is especially so for the expression determining the communication of the agents, which is only implicitly evaluated (and hence selected for) via the effect its communication has on others (and itself). Each agent has a fairly small population of such models (in this case 40). This population of expressions is generated according to the specified language at random. In subsequent generations the population of expressions is developed by a genetic programming [21] algorithm with a lot of propagation and only a little cross-over. The formal language that these expressions are examples of is quite expressive. The primitive nodes and terminals allowed are shown in figure 3. It includes: logical operators, arithmetic, stochastic elements, self-referential operations, listening operations, elements to copy the action of others, statistical summaries of past numbers attending, operations for looking back in time, comparisons and the quote operator.

The reasons for adopting this particular structure for agent cognition is basically that it implements a version of rationality that is credible and bounded but also open-ended and has mechanisms for the expression of complex social distinctions and interaction. In these respects it can be seen as a step towards implementing the ‘model social agent’ described in [6]. For the purposes of this paper the most important aspects are: that the agent constructs its expressions out of previous expressions; that its space of expressions is open-ended allowing for a wide variety of possibilities to be developed; that it has no chance of finding the optimal expressions; and that it is as free from ‘a priori’ design restrictions as is practical and compatible with it having a bounded rationality. This agent architecture and the rationale for its structure is described in more detail in [16, 15].

Communication
Each agent can communicate with any of the others once a week, immediately before they all decide whether to go to the bar or not. The communication is determined by the evaluation of the talk expression and is usually either ‘true’ or ‘false’. The presence of a quoting operator (quote) in the formal language of the talk expression allows subtrees of the talk expression to be the content of the message. If a quote node is reached in the evaluation of the talk expression then the contents of the subtree are passed down verbatim rather than evaluated. If a quoted tree is returned as the result of an evaluation of the talk expression then this is the message that is communicated.

The content of the messages can be used by agents by way of the saidBy and saidByLast nodes in the action and talk expressions. If ‘listening’ is enabled then other agents can use the message in its evaluation of its expressions – if the message is just composed of a boolean value then the saidBy node is just evaluated as this value, but if it is a more complex expression (as a result of a quote node in the sending agents talk expression) then the whole expression will be substituted instead of the saidBy (or saidByLast) node and evaluated as such. The agent can use the output of its own messages by use of other nodes (IPredictedLastWeek and ISaidYesterDay). If ‘imitation’ is enabled then other agents can introduce any message (which is not a mere boolean value) into their own (action) gene pool, this would correspond to agents taking the message as a suggestion for an expression to determine their own action. In subsequent generation this expression can be crossed with other expressions in its population of constructs.

The Results
In figure 6 and figure 7 the attendance patterns of the agents during the eight runs are displayed. The most obvious feature is the difference between the patterns under the crowd-avoiding and friendly runs; under the crowd-avoiding scheme attendance appears far more stochastic compared to those under the friendly scheme where there is obvious coordination. This is unsurprising given that the crowd-avoiding utility scheme encourages the competitive discoordination of behaviour whilst there is a considerable advantage to (at least somewhat) coordinating action with ones ‘friends’ under the friendly scheme.

Comments
The simulation exhibits most of the effects listed above (in the section previous to the description of the model set-up). This is, of course, unsurprising since I have been using the model to hone my intuitions on the topic; the ideas about social embeddedness and the model have themselves co-developed. In particular:
• the expressions that the agents develop resemble constructs rather than models, in that they are opportunistic, they do not reflect their social reality but rather constitute it;
• the constructs can appear highly arbitrary – it can take a great deal of work to unravel them if one attempts to explicitly trace the complex networks of causation (see the examples in the case studies above);
• the agents do frequently use information about the communication and actions of others in stead of attempting to explicitly predict their environment – this is partly confirmed by a general analysis of the general distribution of primitive types in the expressions chosen and developed by agents in figure 16 (the categories the primitives are collected into are fairly self explanatory);
• the agents do specialise as they co-develop their strategies – this is not so apparent from the above but is examined in greater depth elsewhere [14];（essay代写)

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。
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（essay代写)