0411106 20240103182302.220060210235959.9 1-894145-17-8 Waterloo Carleton Scientific 2003 16 s. 305-320BernardJ.-M.SeidenfeldT.ZaffalonM. multidimensional distributions approximations conditional independence cav_un_auth*0101118 Jiroušek Radim UTIA-B Department of Decision Making Theory Ústav teorie informace a automatizace AV ČR, v. v. i. 12A IAA2075302 GA AV ČR cav_un_auth*0001801 CEZ:AV0Z1075907 This paper proposes application of compositional models for approximating multidimensional probability distributions with the help of models defined by a reasonable number of parameters. In addition to a theoretical background, a heuristic algorithm solving one part of a model learning process is presented. Its basic idea, construction of an approximation exploiting informational content of given low-dimensional distributions in a maximal possible way, was proposed originally by Albert Perez. cav_un_auth*0213039 International Symposium on Imprecise Probabilities and Their Applications /3./ Lugano CH 14.07.2003-17.07.2003 BA MTR http://hdl.handle.net/11104/0131193 UTIA-B 20030093 2003 2003 Waterloo Carleton Scientific 1-894145-17-8 Proceedings of the Third International Symposium on Imprecise Probabilities and Their Applications 305 320 Bernard J.-M. 340 Seidenfeld T. 340 Zaffalon M. 340