On the Foundations of Dynamic Games and Probability - Decision Making in Stochastic Extensive Form

Explore the theoretical foundations of dynamic games and probability through this research seminar that addresses the lack of unified decision-theoretic understanding in complex dynamics and uncertainty within game theory. Learn how the speaker combines decision tree theory with Harsanyian exogenous uncertainty concepts to introduce stochastic decision forests, which generalize classical theory to build consistent models of game "rules" through stochastic extensive forms. Discover how well-posed theory based on stochastic processes emerges when the time half-axis is essentially well-ordered, and examine why this condition proves insufficient for many practical applications. Understand the development of a relaxed game-theoretic model using "extensive form characteristics" and see how the problem of instantaneous reaction and information is tackled through vertically extended continuous time with suitable stochastic analysis. Gain insights into how this theoretical framework naturally accommodates information sets, subgames, and equilibrium concepts while applying to stochastic differential and timing games, ultimately addressing open issues in the current literature.