KIT | KIT-Bibliothek | Impressum | Datenschutz

Hierarchical reinforcement learning with multi-step actions

Schoknecht, Ralf

Abstract:

In recent years hierarchical concepts of temporal abstraction
have been integrated in the reinforcement learning framework to
improve scalability. However, existing approaches are limited to
domains for which a decomposition in subtasks is known a priori.
In this paper we propose the concept of multi-step actions on
different time scales in one single action set. It is suited for
learning optimal policies in unstructured domains where a
decomposition is not known in advance or does not exist at all.
At the same time this approach enables learning at multiple
levels of temporal abstraction. Thus, multi-step actions offer
the possibility to obtain faster learning algorithms for
unstructured domains.

Open Access Logo


Volltext §
DOI: 10.5445/IR/30652001
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Logik, Komplexität und Deduktionssysteme (ILKD)
Publikationstyp Buchaufsatz
Publikationsjahr 2001
Sprache Englisch
Identifikator urn:nbn:de:swb:90-AAA306520018
KITopen-ID: 30652001
Erscheinungsvermerk In: Proceedings. Workshop on Hierarchy and Memory in Reinforcement Learning, 18th International Conference on Machine Learning, Williams College, Williamstown, Mass. 2001 [online].
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page