Dynamic Mixture Ratio Model

0524580 - ÚTIA 2021 RIV US eng C - Conference Paper (international conference)
Ruman, Marko - Kárný, Miroslav
Dynamic Mixture Ratio Model.
The Institute of Electrical and Electronics Engineers, Inc., 2020. In: Proceedings of the 2019 International Conference on Control, Artificial Intelligence, Robotics & Optimization (ICCAIRO). Piscataway: IEEE, 2020, s. 92-99, č. článku 19510399. ISBN 978-1-7281-3573-1.
[International Conference on Control, Artificial Intelligence, Robotics & Optimization ICCAIRO 2019. Athens (GR), 08.12.2019-10.12.2019]
R&D Projects: GA MŠMT(CZ) LTC18075
Grant - others:The European Cooperation in Science and Technology (COST)(XE) CA16228
Institutional support: RVO:67985556
Keywords : dynamic systems * mixture models * Bayesian learning * mixture ratio
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
http://library.utia.cas.cz/separaty/2020/AS/karny-0524580.pdf

Finite mixtures of probability densities with components from exponential family serve as flexible parametric models of high-dimensional systems. However, with a few specialized exceptions, these dynamic models assume data-independent weights of mixture components. Their use is illogical and restricts the modeling applicability. The requirement for closeness with respect to conditioning, the basic learning operation, leads to a novel class of models: the mixture ratios. The paper justified them and shows their ability to model truly dynamic systems.
Permanent Link: http://hdl.handle.net/11104/0308930