Claudio Cioffi-Revilla (Director, Center for Social Complexity, Krasnow Institute for Advanced Study, George Mason University)
Agent-based models of coupled socio-natural systems are increasingly useful for theorizing about and analyzing real-world complex systems. Typically, such spatial simulation models include three classes of interdependent components: ecosystems, social systems, and artificial (human-made) systems. The latter serve an adaptive function as interface or buffer between the first two, in the sense of Simon’s Conjecture. An important aspect of these simulation models concerns the in silico occurrence of events, be they natural, human, or technological in origin. Simple event processes are generally those associated with Poisson-like or nearly-Poisson processes, whereas more complex processes include less simple distributions, such as power laws. In this talk I will highlight some of the challenges involved for validation and testing, as well as a possible framework for comparing similarities and differences between simple and complex processes in simulation systems.
This talk is based on research experiences with computational agent-based models, funded by NSF, DARPA, and ONR.
SFI Host: Paula Sabloff