Model Approximation for HEXQ Hierarchical Reinforcement Learning

Hengst, Bernhard

doi:10.1007/978-3-540-30115-8_16

Model Approximation for HEXQ Hierarchical Reinforcement Learning

Bernhard Hengst

ECML-PKDD 2004 pp. 144-155

doi:10.1007/978-3-540-30115-8_16 /ecmlpkdd/2004/hengst2004ecml-model/

Abstract

HEXQ is a reinforcement learning algorithm that discovers hierarchical structure automatically. The generated task hierarchy represents the problem at different levels of abstraction. In this paper we extend HEXQ with heuristics that automatically approximate the structure of the task hierarchy. Construction, learning and execution time, as well as storage requirements of a task hierarchy may be significantly reduced and traded off against solution quality.

PDF ECML-PKDD Semantic Scholar

Cite

Text

Hengst. "Model Approximation for HEXQ Hierarchical Reinforcement Learning." European Conference on Machine Learning, 2004. doi:10.1007/978-3-540-30115-8_16

Markdown

[Hengst. "Model Approximation for HEXQ Hierarchical Reinforcement Learning." European Conference on Machine Learning, 2004.](https://mlanthology.org/ecmlpkdd/2004/hengst2004ecml-model/) doi:10.1007/978-3-540-30115-8_16

BibTeX

@inproceedings{hengst2004ecml-model,
  title     = {{Model Approximation for HEXQ Hierarchical Reinforcement Learning}},
  author    = {Hengst, Bernhard},
  booktitle = {European Conference on Machine Learning},
  year      = {2004},
  pages     = {144-155},
  doi       = {10.1007/978-3-540-30115-8_16},
  url       = {https://mlanthology.org/ecmlpkdd/2004/hengst2004ecml-model/}
}