SubjectsSubjects(version: 945)
Course, academic year 2023/2024
   Login via CAS
Applications of Neural Networks Theory - NAIL013
Title: Aplikace teorie neuronových sítí
Guaranteed by: Department of Theoretical Computer Science and Mathematical Logic (32-KTIML)
Faculty: Faculty of Mathematics and Physics
Actual: from 2016
Semester: summer
E-Credits: 3
Hours per week, examination: summer s.:2/0, Ex [HT]
Capacity: unlimited
Min. number of students: unlimited
4EU+: no
Virtual mobility / capacity: no
State of the course: taught
Language: English, Czech
Teaching methods: full-time
Teaching methods: full-time
Guarantor: doc. RNDr. Iveta Mrázová, CSc.
Class: Informatika Mgr. - Teoretická informatika
Classification: Informatics > Theoretical Computer Science
Annotation -
Last update: RNDr. Filip Zavoral, Ph.D. (03.04.2001)
The course is focused on deeper understanding of the properties and the function of selected models of neural networks - robustness, generalization abilities, etc. Several principles important for the application of neural networks for solving practical tasks will be explained in detail. The discussed application areas include natural speech processing, image processing, robotics, etc.
Aim of the course - Czech
Last update: T_KTI (26.05.2008)

Rozebrat a naučit aplikace neuronových sítí

Course completion requirements -
Last update: RNDr. František Mráz, CSc. (19.02.2024)

In an accompanying Moodle course, there will be published (one) project-oriented assignment along with its working schedule and grading scheme. Each phase of the assignment solution will have a deadline till which it should be submitted for grading. Late submissions will be penalized by a 1% deduction from the overall grading score for each started week of the delay. The completed assignment will count up to 55% of the final score for the exam. The exam at the end of the term will add up to the remaining 45% of the final score. The following table gives the final grade according to the achieved score:

grade 1 grade 2 grade 3 failure
100%–86% 85%–71% 71%–56% less than 56%

Literature -
Last update: doc. RNDr. Iveta Mrázová, CSc. (03.11.2019)

Abu-Mostafa Y. S., Magdon-Ismail M., Lin H.-T.: Learning From Data: A Short Course, AMLbook.com, 2012

Goldberg D.E.: Genetic Algorithms in Search, Optimization, and Machine Learning. Addison Wesley, 1990

Haykin S.: Neural Networks and Learning Machines, 3rd Edition, Pearson, 2009

Kosko B.: Neural Networks for Signal Processing. Prentice Hall, 1992

Syllabus -
Last update: RNDr. František Mráz, CSc. (05.05.2015)

1. Introduction to the area of adaptive and learning systems

  • Adaptation and learning, formal description of patterns, selection and ordering of training patterns.
  • Methods minimizing the loss criterion (Bayesian decision rule, k-nearest neighbor rule, cluster analysis).
  • Applications of classical learning classifiers (in image recognition, speech processing, control).

2. Artificial neural networks and their application

  • A brief recapitulation of selected neural network paradigms (feed-forward neural networks of the back-propagation type, Hopfield networks, Kohonen self-organizing maps, deep neural networks).
  • Applications of neural networks - among others in natural language processing, modeling of financial systems, multimedia data processing, robotics and time series prediction.

3. Application of genetic algorithms in the area of neural networks

  • Application of multi-layered neural networks of the back-propagation type in the evaluation of the fitness functions for genetic algorithms.
  • Optimization of the architecture of neural networks by means of genetic algorithms.

 
Charles University | Information system of Charles University | http://www.cuni.cz/UKEN-329.html