Learning to Plan in Complex Domains

Abstract

Most real-world domains involve large search spaces with large amounts of subgoal interaction. Problems in these domains usually require finding a solution that best meets a set of goals and constraints. Finding such a solution often requires optimizing several real-valued performance measures, as well as satisfying a set of Boolean constraints. Most learning planners have concentrated on problems completely defined by Boolean constraints and have ignored real-valued performance measures. These problems have had low subgoal interaction and small search spaces. We introduce a new general learning problem solver for complex real-world problems.

Cite

Text

Rudy and Kibler. "Learning to Plan in Complex Domains." International Conference on Machine Learning, 1989. doi:10.1016/B978-1-55860-036-2.50051-5

Markdown

[Rudy and Kibler. "Learning to Plan in Complex Domains." International Conference on Machine Learning, 1989.](https://mlanthology.org/icml/1989/rudy1989icml-learning/) doi:10.1016/B978-1-55860-036-2.50051-5

BibTeX

@inproceedings{rudy1989icml-learning,
  title     = {{Learning to Plan in Complex Domains}},
  author    = {Rudy, David and Kibler, Dennis F.},
  booktitle = {International Conference on Machine Learning},
  year      = {1989},
  pages     = {180-182},
  doi       = {10.1016/B978-1-55860-036-2.50051-5},
  url       = {https://mlanthology.org/icml/1989/rudy1989icml-learning/}
}