Premium
Learning for non‐stationary Dirichlet processes
Author(s) -
Quinn A.,
Kárný M.
Publication year - 2007
Publication title -
international journal of adaptive control and signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.73
H-Index - 66
eISSN - 1099-1115
pISSN - 0890-6327
DOI - 10.1002/acs.949
Subject(s) - forgetting , parametric statistics , robustness (evolution) , dirichlet distribution , computer science , dirichlet process , context (archaeology) , parametric model , mathematics , artificial intelligence , mathematical optimization , algorithm , statistics , inference , mathematical analysis , paleontology , philosophy , linguistics , biochemistry , chemistry , biology , gene , boundary value problem
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 on‐line learning of F using i.i.d. 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. Copyright © 2007 John Wiley & Sons, Ltd.