Learning Subjective Representations for Planning
Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence. Pages 889-893.
Abstract
Planning involves using a model of an agent’s actions to find a sequence of decisions which achieve a desired goal. It is usually assumed that the models are given, and such models often require expert knowledge of the domain. This paper explores subjective representations for planning that are learned directly from agent observations and actions (requiring no initial domain knowledge). A non-linear embedding technique called Action Respecting Embedding is used to construct such a representation. It is then shown how to extract the effects of the agent’s actions as operators in this learned representation. Finally, the learned representation and operators are combined with search to find sequences of actions that achieve given goals. The efficacy of this technique is demonstrated in a challenging robot-vision-inspired image domain.
@inproceedings{ijcai2005p218,
title = {Learning Subjective Representations for Planning},
author = {Wilkinson, Dana F. and Bowling, Michael H. and Ghodsi, Ali},
booktitle = {Proceedings of the Nineteenth International Joint Conference on
Artificial Intelligence, {IJCAI-05}},
publisher = {International Joint Conferences on Artificial Intelligence},
pages = {889--893},
year = {2005},
url = {https://ijcai.org/proceedings/2005/218},
}
TY - CPAPER TI - Learning Subjective Representations for Planning AU - Wilkinson, Dana F. AU - Bowling, Michael H. AU - Ghodsi, Ali T2 - Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence, IJCAI-05 PB - International Joint Conferences on Artificial Intelligence PY - 2005 SP - 889 EP - 893 UR - https://ijcai.org/proceedings/2005/218 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
