0085844 20240103184432.320070925235959.9 10 s. cav_un_epca*00858433-540-74693-5Part I.SL 1 - Theoretical Computer Science and General Issues129-138BerlinSpringer2007Marques de SáJ.AlexandreL. A.DuchW.MandicD. neural networks probabilistic approach distribution mixtures cav_un_auth*0101091 Grim Jiří UTIA-B Department of Pattern Recognition Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0230019 Hora Jan UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. 507752 XE EC cav_un_auth*0200689 1M0572 GA MŠk cav_un_auth*0001814 2C06019 GA MŠk CZ cav_un_auth*0216518 GA102/07/1594 GA ČR cav_un_auth*0228611 CEZ:AV0Z10750506 Considering the probabilistic approach to neural networks in the framework of statistical pattern recognition we assume approximation of class-conditional probability distributions by finite mixtures of product components. The mixture components can be interpreted as probabilistic neurons in neurophysiological terms and, in this respect, the fixed probabilistic description becomes conflicting with the well known short-term dynamic properties of biological neurons. We show that some parameters of PNN can be ``released'' for the sake of dynamic processes without destroying the statistically correct decision making. In particular, we can iteratively adapt the mixture component weights or modify the input pattern in order to facilitate the correct recognition. Rekurentní bayesovské odvozování v pravděpodobnostních neuronových sítích cav_un_auth*0230020 International Conference on Artificial Neural Networks /17./ Porto 09.09.2007-13.09.2007 PT 2008 BD http://hdl.handle.net/11104/0148268 2007 cav_un_epca*0085843 Artificial Neural Networks - ICANN 2007 Part I. 3-540-74693-5 129 138 Berlin Springer 2007 Lecture Notes in Computer Scinece 4669 SL 1 - Theoretical Computer Science and General Issues Marques de Sá J. 340 Alexandre L. A. 340 Duch W. 340 Mandic D. 340