Bi-Level Agent Hierarchies (BLAH): A Double-Layered Formalization for Learning and Communication in Multi-Agent Systems
Contributor(s)The Pennsylvania State University CiteSeerX Archives
Full recordShow full item record
AbstractSimulations have become a great tool for research in the natural sciences. However their potential has not been reached in the social sciences. This is in part due to the difficulty in simulating human decision making. Recent advances in neo-classical decision making have defined specific differences between the decision making capabilities of rational agents and humans as well as speculations into the cause. Presented in later sections is a proposition for the formalization of a model for simulating human decision making as well as a method for verifying the method. The flip-side of the social simulation coin requires an examination of systemic pressures and conditions that affect individual agents. In order to formally model such systemic factors, various social metrics must be devised. Specifically, we will go on to examine the issue of simulated social influence. The trouble with metrics that seek to approximate influence across a network is that such metrics are potentially two-sided: agent a might have a high degree of potential influence which a does not actually use. Both of these metrics can be interesting and useful.