Approximate Policy Iteration using Large-Margin Classifiers
Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence. Pages 1432-1433.
Abstract
We present an approximate policy iteration algorithm that uses rollouts to estimate the value of each action under a given policy in a subset of states and a classifier to generalize and learn the improved policy over the entire state space. Using a multiclass support vector machine as the classifier, we obtained successful results on the inverted pendulum and the bicycle balancing and riding domains.
Michail G. Lagoudakis, and Ronald Parr. 2003. Approximate Policy Iteration using Large-Margin Classifiers. In Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, IJCAI-03. IJCAI, 1432–1433. https://ijcai.org/proceedings/2003/226
Lagoudakis, M. G., & Parr, R. (2003). Approximate Policy Iteration using Large-Margin Classifiers. In Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, IJCAI-03 (pp. 1432–1433). IJCAI. https://ijcai.org/proceedings/2003/226
@inproceedings{ijcai2003p226,
title = {Approximate Policy Iteration using Large-Margin Classifiers},
author = {Lagoudakis, Michail G. and Parr, Ronald},
booktitle = {Proceedings of the Eighteenth International Joint Conference on
Artificial Intelligence, {IJCAI-03}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Georg Gottlob},
pages = {1432--1433},
year = {2003},
url = {https://ijcai.org/proceedings/2003/226},
}
TY - CPAPER TI - Approximate Policy Iteration using Large-Margin Classifiers AU - Lagoudakis, Michail G. AU - Parr, Ronald T2 - Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, IJCAI-03 PB - International Joint Conferences on Artificial Intelligence PY - 2003 SP - 1432 EP - 1433 UR - https://ijcai.org/proceedings/2003/226 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
