Game Theory Overview

Principles

Definition

Game theory is a formal study of strategic interactions among rational decision-makers. It employs mathematical models to analyze situations where the outcome for each participant depends on the choices made by all involved, aiming to predict equilibrium outcomes and guide optimal decision-making.

Ontological type

Core Concepts

Solution Concepts

Economic Applications

Key developments

Key Figures

Equilibrium Foundations era

Mancur Olson[1] was associated with Princeton University[3] and the University of Maryland, College Park[4] during this era. Olson[1] contributed foundational insights in The Logic of Collective Action: Public Goods and the Theory of Groups[7], clarifying why individuals may not contribute to public goods and how group dynamics shape outcomes, thereby informing how equilibrium-like behavior emerges in collective decision contexts. Amartya Sen[2] worked with Harvard University[5] and the Massachusetts Institute of Technology[6] during this era. Sen[2] contributed to the formal study of collective choice and welfare through Collective Choice and Social Welfare[8], developing frameworks for evaluating aggregated welfare and decision rules that reinforced the role of information and aggregation in equilibrium analyses.

Dynamic Evolution and Networks era

Klaus M. Schmidt [1] is a key figure in the Dynamic Evolution and Networks era, with affiliations at University College London [3] and Goethe University Frankfurt [4] during this period. Schmidt [1] co-authored the 1999 paper A Theory of Fairness, Competition, and Cooperation [7], which articulated how fairness considerations can influence strategic interaction and foster cooperation, a framing that helped explain how dynamic adaptation unfolds in evolving networks. Ernst Fehr [2] is another central figure, with appointments at Harvard University [5] and California Institute of Technology [6] during this era. Fehr [2] contributed by co-authoring the same 1999 work [7] that highlighted fairness preferences and reputational dynamics in cooperation, and this line of research catalyzed experimental and theoretical work linking reciprocity and network structure to dynamic evolution.

Learning-Driven Multi-Agent Theory era

David Silver [1], affiliated with Stanford University [3] and the University of Washington [4], is a central figure in data-driven multi-agent learning. David Silver [1]'s landmark contribution is the paper Mastering the game of Go without human knowledge [7], which demonstrated self-play and deep reinforcement learning at scale and showed why data-driven learning is pivotal for scalable multi-agent coordination. Thore Graepel [2], affiliated with University College London [5] and the University of Oxford [6], is a central figure in data-driven multi-agent learning. As a co-author of Mastering the game of Go without human knowledge [7], Graepel [2] helped demonstrate scalable data-driven reinforcement learning and self-play, underscoring the era's shift toward empirical coordination.