Techniques for Analysis and Calibration of Multi-agent Simulations

In this paper we present analysis and calibration techniques that exploit knowledge about a multi agent society in order to calibrate the system parameters of a corresponding society simulation model. The techniques address typical problems of multi agent simulation calibration like the vast amount of parameters that need to be calibrated, the complex parameter dependencies due to interactions between the simulated agents and the generally enormous computational cost of running a multi agent simulation. 1 Motivation Multi-Agent Simulation forms a useful tool for understanding and designing societies. As in standard simulation, existing, planned or hypothetic systems are mapped to models. Multi-agent simulation is special as agents and societies in the original are explicitly represented in the model with their autonomy, individual goals, etc. This form of micro simulation provides several advantages not only compared to macro models, but also to other individual- or process based parad...

• Type: Conference
• Year: 2004
• Where: ESAW
• Authors: Manuel Fehler, Franziska Klügl, Frank Puppe
• Keywords: ESAW 2004, Multi Agent Simulation, Multi-agent Simulation, Simulation