Between MDPs and semi-MDPs: a framework for temporal abstraction in reinforcement learning
Artificial Intelligence
Communication-based decomposition mechanisms for decentralized MDPs
Journal of Artificial Intelligence Research
Emotions to control agent deliberation
Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems: volume 1 - Volume 1
ESCAPES: evacuation simulation with children, authorities, parents, emotions, and social comparison
The 10th International Conference on Autonomous Agents and Multiagent Systems - Volume 2
Towards "live" synthetic populations for large-scale realistic multiagent simulations
Proceedings of the 2013 international conference on Autonomous agents and multi-agent systems
Hi-index | 0.00 |
In this paper we describe a multiagent simulation model of human behavior in the aftermath of a hypothetical, large-scale, human-initiated crisis in the center of Washington D.C. Prior studies of this scenario have focused on modeling the physical effects of the attack, such as thermal and blast effects, prompt radiation, and fallout. Casualty and mortality estimates have been obtained by assuming a spatially static human population, ignoring human behavioral response to the event. We build a simulation of a behaving human population and its interaction with various interdependent infrastructures, to try to understand how human response to such an event would change outcomes, and also how modeling this response would enable us to develop new perspectives on planning for this event. Here we present details of the simulation, focusing on the agent design and multiagent interaction, and present initial results on how rapid restoration of communication could alter behavior beneficially.