Applicability of the Future State Maximization Paradigm to Agent-Based Modeling: A Case Study on the Emergence of Socially Sub-Optimal Mobility Behavior

Novel developments in artificial intelligence excel regard to the abilities of rule-based agent-based models (ABMs), but are still limited their representation bounded rationality. The future state maximization (FSX) paradigm presents a promising methodology for describing intelligent behavior agents. FSX agents explore space using “walkers” as virtual entities probing possible states. Recent studies have demonstrated applicability modeling cooperative individuals. Applied ABMs, principle should also represent non-cooperative behavior: example, microscopic traffic modeling, there is need model that do not fully adhere rules. To examine arising from FSX, we developed road section populated by agent-cars endowed with an augmented decision making algorithm. Simulation experiments were conducted four scenarios various settings. A sensitivity analysis showed cooperation among was result balance between exploration and exploitation. We our reproduced several patterns observed models. acting according can stop cooperating. concluded be useful studying irrational certain settings, it suitable ABMs general.