HEXQ

From HandWiki
Revision as of 19:32, 6 March 2021 by imported>JTerm (update)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

HEXQ is a reinforcement learning algorithm created by Bernhard Hengst, which attempts to solve a Markov Decision Process by decomposing it hierarchically.

Bernhard Hengst (2002). Discovering Hierarchy in Reinforcement Learning with HEXQ. http://citeseer.ist.psu.edu/hengst02discovering.html.