Constructing Abstraction Hierarchies Using a Skill-Symbol Loop

Abstract

We describe a framework for building abstraction hierarchies whereby an agent alternates skill- and representation-construction phases to construct a sequence of increasingly abstract Markov decision processes. Our formulation builds on recent results showing that the appropriate abstract representation of a problem is specified by the agent's skills. We describe how such a hierarchy can be used for fast planning, and illustrate the construction of an appropriate hierarchy for the Taxi domain. PDF

Cite

Text

Konidaris. "Constructing Abstraction Hierarchies Using a Skill-Symbol Loop." International Joint Conference on Artificial Intelligence, 2016.

Markdown

[Konidaris. "Constructing Abstraction Hierarchies Using a Skill-Symbol Loop." International Joint Conference on Artificial Intelligence, 2016.](https://mlanthology.org/ijcai/2016/konidaris2016ijcai-constructing/)

BibTeX

@inproceedings{konidaris2016ijcai-constructing,
  title     = {{Constructing Abstraction Hierarchies Using a Skill-Symbol Loop}},
  author    = {Konidaris, George Dimitri},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2016},
  pages     = {1648-1654},
  url       = {https://mlanthology.org/ijcai/2016/konidaris2016ijcai-constructing/}
}