Bayesian stopping rule in discrete parameter space with multiple local maxima

0503809 - ÚTIA 2020 RIV CZ eng J - Journal Article
Kárný, Miroslav
Bayesian stopping rule in discrete parameter space with multiple local maxima.
Kybernetika. Roč. 55, č. 1 (2019), s. 1-11. ISSN 0023-5954
R&D Projects: GA ČR GA16-09848S
Institutional support: RVO:67985556
Keywords : Bayesian estimation * global maximum * model structure
OECD category: Statistics and probability
Impact factor: 0.664, year: 2019
Method of publishing: Open access
http://library.utia.cas.cz/separaty/2019/AS/karny-0503809.pdf https://www.kybernetika.cz/content/2019/1/1

The paper presents the stopping rule for random search for Bayesian model-structure estimation by maximising the likelihood function. The inspected maximisation uses random restarts to cope with local maxima in discrete space. The stopping rule, suitable for any maximisation of this type, exploits the probability of finding global maximum implied by the number of local maxima already found. It stops the search when this probability crosses a given threshold. The inspected case represents an important example of the search in a huge space of hypotheses so common in artificial intelligence, machine learning and computer science.
Permanent Link: http://hdl.handle.net/11104/0295600