考虑排污权交易的有限理性企业博弈模型研究

在有限理性基础上建立寡头垄断博弈模型,将其引入到排污权市场中,同时考虑了生产成本、污染治理成本、排污权交易价格的影响,使该模型更符合实际。将不同理性层次,不同结构的非线性成本函数,不完全信息等因素引入到博弈模型中,对改进后的模型的演化过程进行分析,找出博弈均衡点,并分析其稳定性。由于具有有限理性的双寡头Cournot博弈模型会产生丰富的动力学行为,因此对其进行数据模拟后并对混沌现象进行分析。在此基础上,运用延迟反馈控制法对寡头垄断博弈模型的混沌控制进行了解析分析,结果表明选择合适的控制因子可使模型稳定在Nash均衡。     

A Comparative Study of Game Theoretic and Evolutionary Models of Bargaining for Software Agents

Most of the existing work in the study of bargaining behavior uses techniques from game theory. Game theoretic models for bargaining assume that players are perfectly rational and that this rationality is common knowledge. However, the perfect rationality assumption does not hold for real-life bargaining scenarios with humans as players, since results from experimental economics show that humans find their way to the best strategy through trial and error, and not typically by means of rational deliberation. Such players are said to be boundedly rational. In playing a game against an opponent with bounded rationality, the most effective strategy of a player is not the equilibrium strategy but the one that is the best reply to the opponents strategy. The evolutionary model provides a means for studying the bargaining behaviour of boundedly rational players. This paper provides a comprehensive comparison of the game theoretic and evolutionary approaches to bargaining by examining their assumptions, goals, and limitations. We then study the implications of these differences from the perspective of the software agent developer.     

信息安全攻防博弈研究

信息安全中攻防对抗的本质可以抽象为攻防双方的策略依存性,防御者所采取的防御策略是否有效,不应该只取决于其自身的行为,还应取决于攻击者和防御系统的策略。通过对信息安全攻防不对称性、不完全信息、合理性、重复博弈等特点的分析,以及博弈模型要素和一般模型的分析,构建了信息安全的攻防模型,并对其博弈过程进行了详尽地研究与分析。最后,建议应从攻防博弈的视角来研究信息安全问题,为信息安全问题的解决提供一种新的思路。     

有限理性模型在游戏智能决策中的研究与应用*

为了让游戏NPC能够学习和模拟玩家在游戏中的策略和行为方式,在基于模型的智能决策方法基础上,结合行为决策理论中的有限理性模型提出了一种新的游戏智能方法.该方法分别从有限理性模型的两个核心原则——有限理性和满意准则来改进过去的方法在感知和决策过程中所面对的问题,从而使得游戏NPC行为决策方式更加人性化.最后,通过在Starcraft平台上与其他方法的对抗性实验来进一步验证该方法的优势.     

合作博弈的连续蚁群算法求解

在合作博弈的理论研究中,经典的合作博弈解概念在求解过程中没有体现出局中人的有限理性和互动博弈行为。而在现实博弈环境中,联盟的分配方案更多是通过局中人间理性互动与策略博弈形成的。引入理性因子和控制因子来描述局中人在博弈过程中的决策行为,建立了考虑互动行为的合作博弈模型,并利用连续蚁群算法对合作博弈进行求解。算例表明该解法可以保证分配方案满足有效性和个体理性,并能快速得到联盟的唯一分配方案。这为合作博弈的求解提供了新的思路与工具。     

动态服务的数据交换模型研究*

数据交换是解决信息孤岛问题的实用方法,但如何实施灵活有效的数据交换一直是研究的难点.基于此,提出了动态服务的数据交换方式来克服原有单一、固定的交换缺点.依据交换系统的现状,首先给出了静态服务库、动态服务库和静态交换系统、动态交换系统的概念;然后提出了动态服务的交换模型,并给出了交换的方法、原则和实现过程.最后,结合具体的应用实例,验证了模型的灵活性和有效性.     

面向电动汽车的分布式储能主动汇聚与协同优化方法研究

分布式储能在支撑原有配置需求的基础上具有一定的闲置容量和闲置时间, 若能汇聚这部分闲置储能形成规模化的储能集群, 将能为电网提供多类型辅助服务. 以电动汽车(EV)这一特殊分布式储能为例, 研究考虑个体有限理性特征下的微电网运营商(MGO)对主动电动汽车的汇聚方案: 首先, 建立含主动EV的系统架构, 设计了微电网营运商与主动EV之间的互动流程; 然后, 基于EV的实际出行需求, 利用灵活充放电范围理论建立统一化的EV约束模型; 考虑现实EV用户的有限理性特征, 采用前景理论对其进行有限理性建模; 最后, 将MGO和主动EV之间的互动过程建立为斯塔克伯格博弈(Stackelberg game)模型并验证了博弈均衡解的存在性和唯一性. 算例基于一个商/居型并网微网, 验证了所提模型能有效汇聚分布式EV储能, 促进可再生能源发电就地消纳     

P2P-VCG:一种基于博弈论的带宽分配方案

在引入虚拟货币的前提下,目前的P2P系统没有提供有效的带宽分配和计费方案,因此无法在服务开销和用户收益两个方面达到最优.以用户总体收益为研究点,利用博弈论中的VCG(Vickrey-Clarke-groves)算法规划P2P网络中的带宽分配和计费,并且视客户节点为执行带宽分配的主体,可以很好地克服上述问题,这种基于客户节点的带宽分配方案更能反映P2P网络的分布式特性.该带宽分配算法和基于VCG的计费机制达到了近似系统最优和用户激励两个目标,并且可以在实际系统中得到应用     

The influence of user cognition on consumption decision-making from the perspective of bounded rationality

Consumer decision-making is related to the success or failure of enterprises, and products that cater to the cognitive preferences of users have become a focus of current research. Based on the theory of bounded rationality, this paper explores the cognitive process of consumer decision-making. Then, how product shape affects consumption decision-making is analyzed with eye-tracking technology. Finally, the design principle of the product form is further explored. The results demonstrate the following: (1) the perceptual cognition of users has a driving effect on consumption behavior; (2) as a key factor affecting the perceptual cognition of users, product form affects consumer decision-making by influencing the degree of approach motivation; (3) by establishing the mapping relationship between product form elements and user images, the principles of product form design can be more consistent with user image preferences. This study provides useful suggestions for how to increase the purchase behavior of users from the perspective of bounded rationality.     

Being and acting rational [agent design]

Rationality alone is insufficient to specify agent design. Using economic theory, we can program agents to behave in ways that maximize their utility while responding to environmental changes. However, economic models for agents, although general in principle, are typically limited in practice because the value functions that are tractable essentially reduce an agent to acting selfishly. Building a stable social system from a collection of agents motivated by self-serving interests is difficult. Finally, understanding rationality and knowledge requires interdisciplinary results from artificial intelligence, distributed computing, economics and game theory, linguistics, philosophy, and psychology. A complete theory involves semantic models for knowledge, belief, action, uncertainty; bounded rationality and resource-bounded reasoning; commonsense epistemic reasoning; reasoning about mental states; belief revision; and interactions in multiagent systems.     

基于博弈论优化的高效联邦学习方案

随着网络信息技术与互联网的发展,数据的隐私与安全问题亟待解决,联邦学习作为一种新型的分布式隐私保护机器学习技术应运而生。针对在联邦学习过程中存在个人数据信息泄露的隐私安全问题,结合Micali-Rabin随机向量表示技术,基于博弈论提出一种具有隐私保护的高效联邦学习方案。根据博弈论激励机制,构建联邦学习博弈模型,通过设置合适的效用函数和激励机制保证参与者的合理行为偏好,同时结合Micali-Rabin随机向量表示技术设计高效联邦学习方案。基于Pedersen承诺机制实现高效联邦学习的隐私保护,以保证联邦学习各参与者的利益和数据隐私,并且全局达到帕累托最优状态。在数字分类数据集上的实验结果表明,该方案不仅提高联邦学习的通信效率,而且在通信开销和数据精确度之间实现平衡。     

Computational intelligence determines effective rationality

Rationality is a fundamental concept in economics. Most researchers will accept that human beings are not fully rational. Herbert Simon suggested that we are ＂bounded rational＂. However, it is very difficult to quantify ＂bounded rationality＂, and therefore it is difficult to pinpoint its impact to all those economic theories that depend on the assumption of full rationality. Ariel Rubinstein proposed to model bounded rationality by explicitly specifying the decision makers＇ decision-making procedures. This paper takes a computational point of view to Rubinstein＇s approach. From a computational point of view, decision procedures can be encoded in algorithms and heuristics. We argue that, everything else being equal, the effective rationality of an agent is determined by its computational power - we refer to this as the computational intelligence determines effective rationality （CIDER） theory. This is not an attempt to propose a unifying definition of bounded rationality. It is merely a proposal of a computational point of view of bounded rationality. This way of interpreting bounded rationality enables us to （computationally） reason about economic systems when the full rationality assumption is relaxed.     

Game-theoretic axioms for local rationality and bounded knowledge

We present an axiomatic approach for a class of finite, extensive form games of perfect information that makes use of notions like rationality at a node and knowledge at a node. We distinguish between the game theorist's and the players' own theory of the game. The latter is a theory that is sufficient for each player to infer a certain sequence of moves, whereas the former is intended as a justification of such a sequence of moves. While in general the game theorist's theory of the game is not and need not be axiomatized, the players' theory must be an axiomatic one, since we model players as analogous to automatic theorem provers that play the game by inferring (or computing) a sequence of moves. We provide the players with an axiomatic theory sufficient to infer a solution for the game (in our case, the backwards induction equilibrium), and prove its consistency. We then inquire what happens when the theory of the game is augmented with information that a move outside the inferred solution has occurred. We show that a theory that is sufficient for the players to infer a solution and still remains consistent in the face of deviations must be modular. By this we mean that players have distributed knowledge of it. Finally, we show that whenever the theory of the game is group-knowledge (or common knowledge) among the players (i.e., it is the same at each node), a deviation from the solution gives rise to inconsistencies and therefore forces a revision of the theory at later nodes. On the contrary, whenever a theory of the game is modular, a deviation from equilibrium play does not induce a revision of the theory.A former version of this paper was presented at the Center for Rationality and Interactive Decision Theory at the Hebrew University of Jerusalem. A subsequent version has been presented at the Nobel Symposium on Game Theory held in Björkborn, Sweden, in June 1993. We would like to thank Martin Dufwenberg, Itzhak Gilboa, Sergiu Hart, Bart Lipman, Dov Samet, Shmuel Zamir and especially Robert Aumann for many useful comments.     

有限理性下的企业合作竞争的进化博弈分析

在有限理性的条件下,应用进化博弈论中的模仿者动态模型来分析企业合作竞争博弈的演化,用进化稳定策略来描述合作竞争博弈的长期演化趋势。并分别讨论了对称和非对称的合作竞争博弈的进化均衡与稳定性。最后,指出确定性模仿者动态模型的不足,以及当今随机动态系统理论的发展。     

A social-cognitive approach to online game cheating

Drawing upon social cognitive theory, the present study constructed a conceptual model to provide an in-depth understanding of the key psychosocial factors shaping the cheating behavior in online multiplayer games. An online survey approach and structural equation modeling were adopted to investigate the hypothesized relationships and to assess the fitness of the model. The findings of this study provided support for the research model and the social cognitive theory perspective on game cheating behavior. The cheating behavior was found to be influenced by both external and self-generated factors. Specifically, individual gamers’ social environment (peer influence), their own attitude towards game cheating, and evaluation of cheating outcomes played an important role in shaping this problematic behavior in online games. Game cheating self-efficacy was interestingly found to be an insignificant factor for the game cheating behavior. The study fills a gap in the current literature on games and cheating, specifically on the social cognitive mechanisms of the cheating behavior in online multiplayer games.     

有限理性条件下投票过程的元胞自动机模型研究

有限理性理论认为个体的决策能力是有限的,本文建立元胞自动机（CA）模型研究群体决策中有限理性个体的投票过程,给出了CA的演化规则,模拟了相互邻接的个体问的相互作用,观测到了投票过程的一些微观现象。元胞自动机理论将是推动自然科学和社会科学发展的一个有效工具。     

An evolutionary game algorithm for grid resource allocation under bounded rationality

Resource allocation cannot reach equilibrium in one‐off game in grid environment because of the bounded rationality of the users. To address this issue, an evolutionary game algorithm for grid resource allocation is proposed in this paper. The evolutionary game theory is introduced to study the selection process of user strategy from the dynamic viewpoint. Firstly, the problem of multiple users competing for a common resource is formulated as a symmetric game. Secondly, replicated dynamic mechanism is used to produce the evolutionary stable point that leads to a satisfied allocation scenario. Finally, the relationships among the evolutionary stable point, valuation functions, and convergence time are discussed in detail. The results of the experiments show that the proposed evolutionary game algorithm is convergent and generates better utility results compared with the classical game algorithm. Copyright © 2009 John Wiley & Sons, Ltd.     

基于博弈学习的多Agent城市交通协调控制

交通路口中的各Agent之间的协调问题是一个博弈问题。在有限理性的基础上,利用博弈学习思想,构建多智能体（multi-Agent）博弈学习协调算法,利用此学习协调算法对出行者行为分析并修正,实现城市交通路口的畅通,进而达到区域、全局的交通优化。最后通过实例仿真验证其可行性。     

基于社会选择理论的在线服务评价

用户评价标准不一致和偏好不一致导致网络空间中的在线服务之间不具备公正的可比较性,从而用户难以选择到满意的在线服务,因此,提出了基于社会选择理论计算在线服务优劣的排序方法。首先,根据用户给出的用户-服务评价矩阵构建群体偏好矩阵;然后,基于群体偏好矩阵和Kemeny社会选择函数构建0-1整数规划模型;最后,通过求解该模型可得到服务的最优排序结果。该方法聚合个体偏好为群体偏好,决策符合群体大多数人的偏好且与个体偏好保持最大的一致性。通过理论分析和实验验证了该方法的合理性和有效性。实验结果表明,该方法能有效地解决在线服务之间的不可比较性问题,实现在线服务的优劣排序,并可以有效抵制推荐攻击,具有较强的抗操纵性。