Optimistic Simulated Exploration as an Incentive for Real Exploration

Simple item page
dc.creatorDanihelka, Ivo
dc.date2009-03-17
dc.date2009-05-20
dc.date.accessioned2026-07-07T13:16:21Z
dc.date.available2026-07-07T13:16:21Z
dc.descriptionMany reinforcement learning exploration techniques are overly optimistic and try to explore every state. Such exploration is impossible in environments with the unlimited number of states. I propose to use simulated exploration with an optimistic model to discover promising paths for real exploration. This reduces the needs for the real exploration.
dc.descriptionaccepted, noted that the initial path was 217 steps long
dc.identifierhttps://arxiv.org/abs/0903.2972
dc.identifierhttp://arxiv.org/abs/0903.2972
dc.identifierPOSTER 2009
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/230761
dc.subjectMachine Learning
dc.subjectArtificial Intelligence
dc.titleOptimistic Simulated Exploration as an Incentive for Real Exploration
dc.typetext

Files

Collections

arXiv