Problems in Game Theory and Decision Theory and Their Economic Applications

博弈论和决策论中的问题及其经济应用

• 批准号: 8700454
• 负责人: John Harsanyi
• 金额: $ 7.01万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1987
• 资助国家: 美国
• 起止时间: 1987-06-01 至 1989-11-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8700454&HistoricalAwards=false
• 关键词: Problems; Game; Theory; Decision; Their

This project contributes to game theory, decision theory, and utilitarian ethics. The investigator is a very distinguished scholar. His early work helped define these fields and he remains an active contributor in all of these areas. His recent research is partly responsible for a breakthrough in game theory. He is capable of pursuing this ambitious research agenda. The project consists of three parts. The first part builds on a major breakthrough by the investigator in game theory, i. e., the development of a principled systematic theory that defines a unique outcome for any game. This is an important advance because almost any nontrivial noncooperative game will have very many and sometimes even infinitely many significantly dissimilar outcomes. This means that without this new theory specific testable predictions can be derived from game-theoretic models only by imposing various ad hoc restrictions on acceptable solutions. Further work is needed to define the assumptions that underpin this new theory, contrast the theory with other methods used to obtain unique solutions such as stability theory, and make this theory easier for economists to use. The second part demonstrates that utilitarian theories of ethics are not inconsistent with nonutilitarian concepts of moral obigation and moral rights. The investigator justifies a rule that gives precedence to certain individual moral rights over direct social-utility maximization on the grounds that such a rule has enormous social utility in its own right. If this line of research is successful it could provide a way of rationally justifying moral rights and obligations that modern philosophies now can only define as important. The third part continues the development of a general theory of inductive logic based on Bayesian decision theory. This project explores the relationship between rational presumption of truth and probabilities. It extends earlier work on the relationship between the simplicity of empirical hypotheses and their prior probability.

这个项目对博弈论、决策理论和功利主义伦理学有贡献。调查员是一位非常杰出的学者。他早期的工作帮助定义了这些领域，他仍然是所有这些领域的积极贡献者。他最近的研究在一定程度上促成了博弈论的突破。他有能力完成这个雄心勃勃的研究计划。该项目由三部分组成。第一部分建立在研究者在博弈论方面的重大突破之上，也就是说，发展了一个原则性的系统理论，该理论定义了任何游戏的独特结果。这是一个重要的进步，因为几乎任何非平凡的非合作博弈都会有非常多，有时甚至是无限多的显著不同的结果。这意味着，如果没有这个新的理论，特定的可测试的预测只能通过对可接受的解决方案施加各种特别的限制，才能从博弈论模型中得到。需要进一步的工作来定义支撑这一新理论的假设，将该理论与用于获得唯一解的其他方法（如稳定性理论）进行比较，并使该理论更容易为经济学家所用。第二部分论证了功利主义伦理学理论与非功利主义的道德义务和道德权利概念并不矛盾。研究者为一项将某些个人道德权利置于直接社会效用最大化之上的规则辩护，理由是这种规则本身具有巨大的社会效用。如果这条研究路线取得成功，它可以提供一种理性地证明道德权利和义务的方式，现代哲学现在只能将其定义为重要的。第三部分继续在贝叶斯决策理论的基础上发展归纳逻辑的一般理论。这个项目探讨了理性假设真理和概率之间的关系。它扩展了早期关于经验假设的简单性与其先验概率之间关系的研究。