Course: Artificial Neural Networks - Selected Chapters

« Back
Course title Artificial Neural Networks - Selected Chapters
Course code KAM/DUNES
Organizational form of instruction no contact
Level of course Doctoral
Year of study 2
Semester Winter and summer
Number of ECTS credits 20
Language of instruction Czech, English
Status of course Compulsory-optional
Form of instruction Face-to-face
Work placements This is not an internship
Recommended optional programme components None
Lecturer(s)
  • Doležel Petr, prof. Ing. Ph.D.
Course content
The course deals with the most commonly used artificial neural networks paradigms (feedforward multilayer artificial neural network, self-organizing maps, convolutional neural network) including advanced learning algorithms. The properties of the mentioned paradigms will be demonstrated in particular by examples of applications for modeling and control of dynamic systems, but also partly in the processing of audiovisual data, classification and decision making. In addition, students will independently solve a complex problem using an artificial neural network, and the results will be validated on real laboratory device. Main topics: *Feedforward neural network (FFNN) - topology, design, features and qualities, gradient-based learning algorithms, Levenberg-Marquardt learning algorithm, acceleration of learning using special hardware, dynamic systém modelling using FFNN, process control using FFNN *Self-organising maps (SOM) - topology, design, features and qualities, cluster analysis using SOM *Convolutional neural network (CNN) - topology, design, features and qualities, gradient-based learning algorithms for CNN, special hardware for CNN, data compression using CNN, object detection using CNN

Learning activities and teaching methods
Monologic (reading, lecture, briefing), Methods of individual activities, Laboratory work
Learning outcomes
The course deals with the most commonly used artificial neural networks paradigms (feedforward multilayer artificial neural network, self-organizing maps, convolutional neural network) including advanced learning algorithms.

Prerequisites
unspecified

Assessment methods and criteria
Oral examination, Written examination, Home assignment evaluation

The student completes at least 3 consultations during the semester concerning the theoretical content of the course. The student will pass at least 1 consultations concerning the assigned practical problem.
Recommended literature
  • Goodfellow, I.; Bengio, Y.; Courville, A. Deep learning. Cambridge: The MIT Press, 2016. ISBN 978-0-262-03561-3.
  • Haykin, Simon S. Neural networks and learning machines. Upper Saddle River: Prentice Hall, 2009. ISBN 978-0-13-147139-9.
  • KVASNIČKA, V. a kol. Úvod do teórie neurónových sietí.. Bratislava: IRIS, 1997. ISBN 80-88778-30-1.
  • Nguyen, Hung T. A first course in fuzzy and neural control. Boca Raton: Chapman & Hall, 2003. ISBN 1-58488-244-1.
  • VONDRÁK, I. Umělá inteligence a neuronové sítě. Ostrava: VŠB - TU Ostrava, 2009. ISBN 978-8-02-481981-5.


Study plans that include the course
Faculty Study plan (Version) Category of Branch/Specialization Recommended year of study Recommended semester