Learning to Rank for Synthesizing Planning Heuristics
Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence. Pages 3089-3094.
Abstract
We investigate learning heuristics for domain-specific planning. Prior work framed learning a heuristic as an ordinary regression problem. However, in a greedy best-first search, the ordering of states induced by a heuristic is more indicative of the resulting planner's performance than mean squared error. Thus, we instead frame learning a heuristic as a learning to rank problem which we solve using a RankSVM formulation. Additionally, we introduce new methods for computing features that capture temporal interactions in an approximate plan. Our experiments on recent International Planning Competition problems show that the RankSVM learned heuristics outperform both the original heuristics and heuristics learned through ordinary regression.
@inproceedings{ijcai2016p438,
title = {Learning to Rank for Synthesizing Planning Heuristics},
author = {Garrett, Caelan Reed and Kaelbling, Leslie Pack and Lozano-Perez, Tomas},
booktitle = {Proceedings of the Twenty-Fifth International Joint Conference on
Artificial Intelligence, {IJCAI-16}},
publisher = {AAAI Press / International Joint Conferences on Artificial Intelligence},
editor = {Subbarao Kambhampati},
pages = {3089--3094},
year = {2016},
month = {7},
url = {https://ijcai.org/proceedings/2016/438},
}
TY - CPAPER TI - Learning to Rank for Synthesizing Planning Heuristics AU - Garrett, Caelan Reed AU - Kaelbling, Leslie Pack AU - Lozano-Perez, Tomas T2 - Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-16 PB - AAAI Press / International Joint Conferences on Artificial Intelligence PY - 2016 SP - 3089 EP - 3094 UR - https://ijcai.org/proceedings/2016/438 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
