SubjectsSubjects(version: 845)
Course, academic year 2019/2020
   Login via CAS
Seminar of machine learning and modelling I - NAIL099
Title in English: Seminář strojového učení a modelování I
Guaranteed by: Department of Theoretical Computer Science and Mathematical Logic (32-KTIML)
Faculty: Faculty of Mathematics and Physics
Actual: from 2019 to 2019
Semester: winter
E-Credits: 2
Hours per week, examination: winter s.:0/1 C [hours/week]
Capacity: unlimited
Min. number of students: unlimited
State of the course: taught
Language: Czech
Teaching methods: full-time
Additional information:
Note: enabled for web enrollment
Guarantor: doc. RNDr. Ing. Martin Holeňa, CSc.
Class: Informatika Mgr. - volitelný
Classification: Informatics > Informatics, Software Applications, Computer Graphics and Geometry, Database Systems, Didactics of Informatics, Discrete Mathematics, External Subjects, General Subjects, Computer and Formal Linguistics, Optimalization, Programming, Software Engineering, Theoretical Computer Science, Theoretical Computer Science
Annotation -
Last update: T_KTI (10.05.2010)
Seminar of machine learning and modelling is oriented to methods of machine learning and modelling based on data. Work of Mgr. and PhD. students, lectures of researchers and occasionally invided lectures of foreign visitors from this area are presented. We invite also students, which want to refere about some book or paper in area of machine learning or modelling based on data.
Aim of the course -
Last update: doc. RNDr. Ing. Martin Holeňa, CSc. (29.06.2019)

Enable coming into and keeping contacts between master and PhD students from the Faculty of Mathematics and Physics and from the Czech Technical University working on the topics in the areas machine learning and data-driven modelling. Mediate the exchange of experience between them and with scientists from those areas invited to the seminar. Provide them feedback and inspiring ideas.

Course completion requirements -
Last update: doc. RNDr. Ing. Martin Holeňa, CSc. (29.06.2019)

For this course to be credited, it is sufficient to attend all seminars (except excused absence), but during 2 subsequent terms. Hence, for the credit from Seminar of machine learning and modelling I, it has to be attended in a summer term and the subsequent winter term, for the credit from Seminar of machine learning and modelling II, it has to be attended in a winter term and the subsequent summer term.

Literature - Czech
Last update: T_KTI (10.05.2010)

Seminář nemá jednotný seznam literatury, protože diplomové a disertační práce se opírají především o specifickou literaturu k řešenému tématu. Vítáni jsou i zájemci, kteří četli nějakou zajímavou knížku či přehledový článek související se zaměřením semináře a mají chuť o tom poreferovat.

Syllabus -
Last update: T_KTI (10.05.2010)

The following list of themas is about interests of lecturers from previous years, but does not limit future themas. Invited are all themas relevant to machine learning and modelling based on data.

Examples of previous themas:

Learning of ruled from data, learning of boolean anf fuzzy rules.

Association and clasification rules.

Inductive inference, inductive logic programming.

Case-based study, transductive inference.

Statistical learning, PAC learning.

Clustering based on similarity, clustering using self-organisation.

Evolutional learning, evolutional extraction of rules from data.

Genetic algorithms, genetic programing.

Evolutional algorithms based on differential evolution and on estimations of distibution.

Learning of artificial neural nets (with and without supervisor).

Perceptrons and multilayered perceptrons.

Neuron nets with radial basis functions.

Self-organising maps, combining of neuron nets and evolution algorithms.

Clasification and regression using support vector machines.

Hierarchical regression models.

Decision trees for classification and regression.

Combining of decision trees to random forests.

General methods of combining classifiers, ensemble methods.

Fuzzy aggregation of classifiers, fuzzy classification.

General methods of combining regression models, reliability of prediction.

Data visualization, visualization of models constructed from data.

Applications of machine learning methods in physics, chemistry, biology and computer games.

Application of models extracted from data in natural and technical sciences.

Charles University | Information system of Charles University |