000003392 001__ 3392 000003392 005__ 20141118153401.0 000003392 04107 $$acze 000003392 046__ $$k2005-05-09 000003392 100__ $$aVěchet, S. 000003392 24500 $$aNonstationary system control using Q-learning 000003392 24630 $$n11.$$pENGINEERING MECHANICS 2005 000003392 260__ $$bInstitute of Thermomechanics AS CR, v.v.i., Brno 000003392 506__ $$arestricted 000003392 520__ $$2eng$$aQ-learning is the most popular and effective version of Reinforcement Learning algorithms. In this paper we discuss the possibility of control of a nonstationary system by Q-learning. The non-stationary system is represented by simple inverted pendulum simulation model with variable pendulum length. 000003392 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb. 000003392 653__ $$a 000003392 7112_ $$aENGINEERING MECHANICS 2005$$cSvratka (CZ)$$d2005-05-09 / 2005-05-12$$gEM2005 000003392 720__ $$aVěchet, S.$$iKrejsa, J. 000003392 8560_ $$ffischerc@itam.cas.cz 000003392 8564_ $$s222876$$uhttps://invenio.itam.cas.cz/record/3392/files/Vechet1-PT.pdf$$y Original version of the author's contribution as presented on CD, . 000003392 962__ $$r3238 000003392 980__ $$aPAPER