Success, strategy and skill: an experimental study

In many AI settings an agent is comprised of both actionplanning and action-execution components. We examine the relationship between the precision of the execution component, the intelligence of the planning component, and the overall success of the agent. Our motivation lies in determining whether higher execution skill rewards more strategic playing. We present a computational billiards framework in which the interaction between skill and strategy can be experimentally investigated. By comparing the performance of different agents with varying levels of skill and strategic intelligence we show that intelligent planning can contribute most to an agent's success when that agent has imperfect skill. Categories and Subject Descriptors I.2.8 [Artificial Intelligence]: Problem Solving, Control Methods, and Search--Plan execution; I.2.1 [Artificial Intelligence]: Applications and Expert Systems--Games; I.2.11 [Artificial Intelligence]: Multiagent Systems General Terms Experimentation...

• Type: Conference
• Year: 2010
• Where: ATAL
• Authors: Christopher Archibald, Alon Altman, Yoav Shoham
• Keywords: Artificial Intelligence, ATAL 2010, Computational Billiards, Computational Billiards Framework, Intelligent Agents