Corpus ID: 5588386

Discovering Hierarchy in Reinforcement Learning with HEXQ

@inproceedings{Hengst2002DiscoveringHI,
  title={Discovering Hierarchy in Reinforcement Learning with HEXQ},
  author={B. Hengst},
  booktitle={ICML},
  year={2002}
}

• B. Hengst
• Published in ICML 2002
• Computer Science

people.eecs.berkeley.edu

Save to Library

Create Alert

Cite

Launch Research Feed

An open problem in reinforcement learning is discovering hierarchical structure. HEXQ, an algorithm which automatically attempts to decompose and solve a model-free factored MDP hierarchically is described. By searching for aliased Markov sub-space regions based on the state variables the algorithm uses temporal and state abstraction to construct a hierarchy of interlinked smaller MDPs.