Anzeige
Erschienen in:
2013 | OriginalPaper | Buchkapitel
6. Up the Down Staircase: Hierarchical Reinforcement Learning
verfasst von : Alexander Paprotny, Michael Thess
Erschienen in: Realtime Data Mining
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
We address the question of how hierarchical, or multigrid, methods may figure in dynamic programming and reinforcement learning for recommendation engines.
After providing a general introduction, we approach the framework of hierarchical methods from both the historical analytical and algebraic viewpoints; we proceed to devising and justifying approaches to apply hierarchical methods to both the model-based as well as the model-free case. In regard to the latter, we set out from the multigrid reinforcement learning algorithms introduced by Ziv in [Ziv04] and extend these methods to finite-horizon problems.