Efficient Reinforcement Learning with Relocatable Action Models

Abstract

Realistic domains for learning possess regularities that make it possible to generalize experience across related states. This paper explores an environment-modeling framework that rep-resents transitions as state-independent outcomes that are common to all states that share the same type. We analyze a set of novel learning problems that arise in this framework, providing lower and upper bounds. We single out one partic-ular variant of practical interest and provide an efficient algo-rithm and experimental results in both simulated and robotic environments.

Cite

Text

Leffler et al. "Efficient Reinforcement Learning with Relocatable Action Models." AAAI Conference on Artificial Intelligence, 2007.

Markdown

[Leffler et al. "Efficient Reinforcement Learning with Relocatable Action Models." AAAI Conference on Artificial Intelligence, 2007.](https://mlanthology.org/aaai/2007/leffler2007aaai-efficient/)

BibTeX

@inproceedings{leffler2007aaai-efficient,
  title     = {{Efficient Reinforcement Learning with Relocatable Action Models}},
  author    = {Leffler, Bethany R. and Littman, Michael L. and Edmunds, Timothy},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2007},
  pages     = {572-577},
  url       = {https://mlanthology.org/aaai/2007/leffler2007aaai-efficient/}
}