0364820 20240103195601.920111101235959.9 000297000300012 80054779788 10.1016/j.neunet.2011.06.006 8 s. cav_un_epca*02573100893-60802410 (2011)1128-1135Elsevier Stochastic control design Fully probabilistic design Adaptive control Adaptive critic cav_un_auth*0275438 Herzallah R. JO cav_un_auth*0101124 Kárný Miroslav Adaptivní systémy Department of Adaptive Systems AS AS UTIA-B Department of Adaptive Systems Ústav teorie informace a automatizace AV ČR, v. v. i. http://library.utia.cas.cz/separaty/2011/AS/karny-0364820.pdf GA102/08/0567 GA ČR cav_un_auth*0239566 CEZ:AV0Z10750506 Optimal stochastic controller pushes the closed-loop behavior as close as possible to the desired one. The fully probabilistic design (FPD) uses probabilistic description of the desired closed loop and minimizes Kullback–Leibler divergence of the closed-loop description to the desired one. Practical exploitation of the fully probabilistic design control theory continues to be hindered by the computational complexities involved in numerically solving the associated stochastic dynamic programming problem; in particular, very hard multivariate integration and an approximate interpolation of the involved multivariate functions. This paper proposes a new fully probabilistic control algorithm that uses the adaptive critic methods to circumvent the need for explicitly evaluating the optimal value function, thereby dramatically reducing computational requirements. This is a main contribution of this paper. 2012 BC 4 A 4a 20231122134701.4 http://hdl.handle.net/11104/0200201 COMPUTERSCIENCEARTIFICIALINTELLIGENCE|NEUROSCIENCES 2.477 0.318 9.9 0.01073 0.918 5336 110 87 0.835 40.73 3.16 Q2 49.356 56.771 Q3 Q3 2011 Soubory v repozitáři: karny-0364820.pdf 80054779788 SCOPUS 000297000300012 WOS cav_un_epca*0257310 Neural Networks 0893-6080 1879-2782 Roč. 24 č. 10 2011 1128 1135 Elsevier