Time manipulation technique for speeding up reinforcement learning in simulations

Simple item page
dc.creatorKormushev, Petar
dc.creatorNomoto, Kohei
dc.creatorDong, Fangyan
dc.creatorHirota, Kaoru
dc.date2009-03-28
dc.date.accessioned2026-07-07T12:57:37Z
dc.date.available2026-07-07T12:57:37Z
dc.descriptionA technique for speeding up reinforcement learning algorithms by using time manipulation is proposed. It is applicable to failure-avoidance control problems running in a computer simulation. Turning the time of the simulation backwards on failure events is shown to speed up the learning by 260% and improve the state space exploration by 12% on the cart-pole balancing task, compared to the conventional Q-learning and Actor-Critic algorithms.
dc.description12 pages
dc.identifierhttps://arxiv.org/abs/0903.4930
dc.identifierhttp://arxiv.org/abs/0903.4930
dc.identifierInternational Journal of Cybernetics and Information Technologies, vol. 8, no. 1, pp. 12-24, 2008
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/224994
dc.subjectArtificial Intelligence
dc.subjectMachine Learning
dc.subjectRobotics
dc.titleTime manipulation technique for speeding up reinforcement learning in simulations
dc.typetext

Files

Collections

arXiv