Počet záznamů: 1
Probabilistic Tools for Optimization of Classifiers on Large Data Sets
- 1.
SYSNO ASEP 0523714 Druh ASEP A - Abstrakt Zařazení RIV Záznam nebyl označen do RIV Zařazení RIV Není vybrán druh dokumentu Název Probabilistic Tools for Optimization of Classifiers on Large Data Sets Tvůrce(i) Kůrková, Věra (UIVT-O) RID, SAI, ORCID
Sanguineti, M. (IT)Zdroj.dok. ODS 2019. Book of Abstracts. - Genova : AIRO - Italian Operations Research Society, 2019
S. 75-75Poč.str. 1 s. Akce ODS 2019: International Conference on Optimization and Decision Science /49./ Datum konání 04.09.2019 - 07.09.2019 Místo konání Genova Země IT - Itálie Typ akce EUR Jazyk dok. eng - angličtina Země vyd. IT - Itálie Klíč. slova Classification ; Optimization of Computational Models ; Concentration of Measures ; Azuma-Hoeffding Inequality CEP GA18-23827S GA ČR - Grantová agentura ČR Institucionální podpora UIVT-O - RVO:67985807 Anotace The number of classification tasks on a finite domain (representing a set of vectors of features, measurements, or observations) grows exponentially with its size. However, for a given application area relevance of many such tasks might be very low or negligible. A probabilistic framework is introduced, modeling prior knowledge about probabilities that a presence of some features implies a property described by one of the classes. Impact of increasing sizes of domains on correlations between input-output mappings of computational models and randomly-chosen classifiers is analyzed. It is proven that for large domains the correlations are sharply concentrated around their mean values. Probabilistic bounds are derived via implications of the Azuma-Hoeffding Inequality, holding also without the ”naive Bayes assumption”. It is shown that the performance of random classifiers is almost deterministic, in the sense that either a given class of computational models can approximate well almost all tasks or none of them. Consequences for the choice of optimal computational models are derived Pracoviště Ústav informatiky Kontakt Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Rok sběru 2021
Počet záznamů: 1