Learning Recursive Control Programs from Problem Solving
Abstract
In this paper, we propose a new representation for physical control -- teleoreactive logic programs -- along with an interpreter that uses them to achieve goals. In addition, we present a new learning method that acquires recursive forms of these structures from traces of successful problem solving. We report experiments in three different domains that demonstrate the generality of this approach. In closing, we review related work on learning complex skills and discuss directions for future research on this topic.
Cite
Text
Langley and Choi. "Learning Recursive Control Programs from Problem Solving." Journal of Machine Learning Research, 2006.Markdown
[Langley and Choi. "Learning Recursive Control Programs from Problem Solving." Journal of Machine Learning Research, 2006.](https://mlanthology.org/jmlr/2006/langley2006jmlr-learning/)BibTeX
@article{langley2006jmlr-learning,
title = {{Learning Recursive Control Programs from Problem Solving}},
author = {Langley, Pat and Choi, Dongkyu},
journal = {Journal of Machine Learning Research},
year = {2006},
pages = {493-518},
volume = {7},
url = {https://mlanthology.org/jmlr/2006/langley2006jmlr-learning/}
}