IA178 Bayes networks

Fakulta informatiky
jaro 2026

Předmět se v období jaro 2026 nevypisuje.

Rozsah
2/2/0. 3 kr. (plus ukončení). Ukončení: zk.
Vyučováno kontaktně
Vyučující
prof. RNDr. Antonín Kučera, Ph.D. (přednášející)
Garance
prof. RNDr. Antonín Kučera, Ph.D.
Katedra teorie programování – Fakulta informatiky
Kontaktní osoba: prof. RNDr. Antonín Kučera, Ph.D.
Dodavatelské pracoviště: Katedra teorie programování – Fakulta informatiky
Předpoklady
The course assumes familiarity with basic notions of probability theory and elementary mathematical skills.
Omezení zápisu do předmětu
Předmět je otevřen studentům libovolného oboru.
Cíle předmětu
The course focuses on Bayes networks and associated techniques for extracting and encoding knowledge from data. When compared to other representations available for data analysis, the distinguishing feature of Bayes networks is their ability to handle incomplete data sets and learn causal relationships. In conjunction with Bayesian statistical techniques, they facilitate the combination of domain knowledge and data and offer an efficient and principled approach for avoiding the overfitting of data. The course's main goals are to explain the methods for constructing Bayes networks, summarize Bayesian statistical methods for using data to improve these models, and illustrate the graphical modeling approach on real-world examples.
Výstupy z učení
After completing the course, the student will be able to
understand Bayesian networks and their main application domains;
perform statistical inference on a given Bayesian network;
construct Bayesian networks from prior knowledge;
apply methods for handling incomplete data;
understand techniques for learning the probabilities and structure of a Bayesian network;
Osnova
  • Bayesian interpretation of probability.
  • Methods from Bayesian statistics for combining prior knowledge with data.
  • Bayesian networks and their construction.
  • Algorithms for probabilistic inference in Bayesian networks.
  • Methods for learning probabilities in a given Bayesian network.
  • Techniques for handling incomplete data (Monte-Carlo method, Gaussian approximation).
  • Learning the probabilities and structure of a Bayesian network.
  • Bayesian-network techniques and methods for supervised and unsupervised learning.
Literatura
    doporučená literatura
  • JOHNSON, Alicia A.; Miles Q. OTT a Mine DOGUCU. Bayes rules! : an introduction to applied Bayesian modeling. First edition. Boca Raton: CRC Press/Taylor & Francis Group, 2022, xxi, 521. ISBN 9780367255398. info
  • STONE, James V. Bayes' rule : a tutorial introduction to Bayesian analysis. First edition. [Sheffield]: Sebtel Press, 2013, 170 stran. ISBN 9780956372840. info
  • KOLLER, Daphne a Nir FRIEDMAN. Probabilistic graphical models : principles and techniques. Cambridge: MIT Press, 2009, xxxv, 1231. ISBN 9780262013192. info
Výukové metody
Lectures, class discussion, and group projects.
Metody hodnocení
Written test.
Vyučovací jazyk
Angličtina
Další komentáře
Předmět je vyučován každoročně.
Výuka probíhá každý týden.
  • Permalink: https://is.muni.cz/predmet/fi/jaro2026/IA178