. "Learning for Nonstationary Dirichlet Processes" . . "International Journal of Adaptive Control and Signal Processing" . "GB - Spojen\u00E9 kr\u00E1lovstv\u00ED Velk\u00E9 Brit\u00E1nie a Severn\u00EDho Irska" . . "28"^^ . "Learning for Nonstationary Dirichlet Processes"@en . "U\u010Den\u00ED nestacion\u00E1rn\u00EDch Dirichletov\u00FDch proces\u016F"@cs . "10" . "0890-6327" . "[8F3F862C48E6]" . "Learning for Nonstationary Dirichlet Processes"@en . "430687" . . . . . . "The Dirichlet process prior (DPP) is used to model an unknown probability distribution, F: This eliminates the need for parametric model assumptions, providing robustness in problems where there is significant model uncertainty. Two important parametric techniques for learning are extended to this non-parametric context for the first time. These are (i) sequential stopping, which proposes an optimal stopping time for online learning of F using id. sampling; and (ii) stabilized forgetting, which updates the DPP in response to changes in F; but without the need for a formal transition model. In each case, a practical and highly tractable algorithm is revealed, and simulation studies are reported." . . "Quinn, A." . "Learning for Nonstationary Dirichlet Processes" . . "K\u00E1rn\u00FD, Miroslav" . "2"^^ . "The Dirichlet process prior (DPP) is used to model an unknown probability distribution, F: This eliminates the need for parametric model assumptions, providing robustness in problems where there is significant model uncertainty. Two important parametric techniques for learning are extended to this non-parametric context for the first time. These are (i) sequential stopping, which proposes an optimal stopping time for online learning of F using id. sampling; and (ii) stabilized forgetting, which updates the DPP in response to changes in F; but without the need for a formal transition model. In each case, a practical and highly tractable algorithm is revealed, and simulation studies are reported."@en . "RIV/67985556:_____/07:00090636!RIV08-AV0-67985556" . "21" . "1"^^ . . "Nestacion\u00E1rn\u00ED procesy; u\u010Den\u00ED; Dirichletovy procesy; zapom\u00EDn\u00E1n\u00ED"@en . . "Dirichlet\u016Fv proces se pou\u017E\u00EDv\u00E1 pro modelov\u00E1n\u00ED nezn\u00E1m\u00E9ho rozlo\u017Een\u00ED F. To umo\u017E\u0148uje se vyhnout p\u0159edpokladu o jej\u00EDm parametrick\u00E9m rozli\u0161en\u00ED a zaji\u0161\u0165uje odolnost v \u00FAloh\u00E1ch s vysokou neur\u010Ditost\u00ED. \u010Cl\u00E1nek ro\u0161i\u0159uje pou\u017Eit\u00ED dvou technik zn\u00E1m\u00FDch z parametrick\u00E9ho odhadov\u00E1n\u00ED do neparametrick\u00E9ho kontextu. Konkr\u00E9tn\u011B (i) pr\u016Fb\u011B\u017En\u00E9 zastavov\u00E1n\u00ED vol\u00EDc\u00ED okam\u017Eik zastaven\u00ED pr\u016Fb\u011B\u017En\u00E9ho odhadov\u00E1n\u00ED F z nez\u00E1visl\u00FDch vzork\u016F; (ii) stabilizovan\u00E9 zapom\u00EDn\u00E1n\u00ED, kter\u00E9 respektuje pomal\u00E9 \u010Dasov\u00E9 zm\u011Bny F bezjejich podrobn\u011Bj\u0161\u00EDho modelov\u00E1n\u00ED. V obou p\u0159\u00EDpadech je navr\u017Eeno algoritmick\u00E9 \u0159e\u0161en\u00ED a jeho chov\u00E1n\u00ED je ilustrov\u00E1no simula\u010Dn\u00EDmi p\u0159\u00EDklady."@cs . "827;855" . "U\u010Den\u00ED nestacion\u00E1rn\u00EDch Dirichletov\u00FDch proces\u016F"@cs . "RIV/67985556:_____/07:00090636" . . . "P(1ET100750401), Z(AV0Z10750506)" . . . . .