Output details
11 - Computer Science and Informatics
University of Aberdeen
Learning Strategies for Task Delegation in Norm-Governed Environments
<22>This appears in the foremost journal in the field of autonomous agents and multiagent systems. It builds upon a full AAMAS 2011 conference paper http://www.ifaamas.org/Proceedings/aamas2011/papers/A7_G70.pdf (a selective conference with >80% rejection), an ADMI 2012 workshop paper http://dx.doi.org/10.1007/978-3-642-27609-5_9, and an ArgMAS 2011 workshop paper. Using a novel adaptation of decision tree learning that exploits domain ontologies, this is the first attempt to empirically evaluate how information about agents from previous dialogues can be used to build a model of an opponent to be strategically exploited in future encounters.