Output details
11 - Computer Science and Informatics
University of Huddersfield
Action Knowledge Acquisition with Opmaker2
<22>This paper describes and evaluates the first technique to induce efficient representations of PDDL-like actions from training sequences without intermediate state information, large numbers of training examples, or user advice. It was the culmination of work, which included Richardson's PhD thesis, on the automated learning of operator and HTN schema which make up AI planning domain models. The paper is in the post-conference book containing selected papers from the proceedings of ICAART 2009, held in Porto, January 2009. This paper is among the 21 of the 161 submitted papers that were accepted for inclusion: http://link.springer.com/book/10.1007/978-3-642-11819-7.