Game theoretic modeling and analysis : A co-evolutionary, agent-based approach

Abstract

The thesis provides comprehensive treatment on the application of co-evolutionary algorithms to simulate learning and adaptation in agent-based models. This is framed as a simple but complementary alternative to overcome difficulties encountered in analytical and empirical methods e.g. mathematical intractability, limitations in the scope of study, static process of solution discovery and unrealistic assumptions, and to achieve effective modeling that yields meaningful analysis and insights into game theoretic interaction by integrating realistic and dynamic elements into the learning process of individual entities. Through modeling and analysis of repeated games, which find extensive applicability in real world setups, simulated results showed that co-evolutionary, agent-based approaches can discover strategies which are comparable or better than existing ones; and replicate interesting emergent behavior and trends which provide insights into the complexity of collective interaction among diverse strategy types in theoretical and complex situations which often lie beyond their original scope of assumptions.