Multiagent systems: a modern approach to distributed artificial intelligence
Multiagent systems: a modern approach to distributed artificial intelligence
Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence
Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence
Introduction to Multiagent Systems
Introduction to Multiagent Systems
Accident or intention: that is the question (in the Noisy Iterated Prisoner's Dilemma)
AAMAS '06 Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems
The Iterated Prisoners' Dilemma: 20 Years on
The Iterated Prisoners' Dilemma: 20 Years on
Approximating game-theoretic optimal strategies for full-scale poker
IJCAI'03 Proceedings of the 18th international joint conference on Artificial intelligence
Hi-index | 0.00 |
To prevent or alleviate conflicts in multi-agent environments, it is important to distinguish between situations where another agent has misbehaved intentionally and situations where the misbehavior was accidental. One situation where this problem arises is the Noisy Iterated Prisoner's Dilemma, a version of the Iterated Prisoner's Dilemma (IPD) in which there is a nonzero probability that a "cooperate" action will accidentally be changed into a "defect" action and vice versa. Tit-For-Tat and other strategies that do quite well in the ordinary (non-noisy) IPD can do quite badly in the Noisy IPD. This paper presents a technique called symbolic noise detection, for detecting whether anomalies in player's behavior are deliberate or accidental. This idea to use player's deterministic behavior to tell whether an action has been affected by noise. We also present DBS, an algorithm that uses symbolic noise detection in the Noisy IPD. DBS constructs a model of the other agent's deterministic behavior, and watches for any deviation from this model. If the other agent's next action is inconsistent with this model, the inconsistency can be due either to noise or to a genuine change in their behavior; and DBS can often distinguish between two cases by waiting to see whether this inconsistency persists in next few moves. This technique is effective because many IPD players often have clear deterministic patterns of behavior. We entered several different implementations of DBS in the 2005 Iterated Prisoner's Dilemma competition, in Category 2 (noisy environments). Out of the 165 contestants in this category, most of DBS implementations ranked among top ten. The best one ranked third, and it was beaten only by two "master-and-slaves strategy" programs that each had a large number of "slave" programs feeding points to them.