Course detail

Artificial Intelligence

FEKT-NUINAcad. year: 2018/2019

The aim of the course is to deepen knowledges and application of artificial intelligence methods. Artificial intelligence – definition, trends. Artificial neural networks, neural networks paradigms, method of backpropagation learning, Kohonen self-organizing maps, Hopfield network, RCE neural network. Knowledge-based systems, knowledge representation, problem solving, structure and activities of expert systems. Optical information processing resources of artificial inteligence. Intelligent robot.

Language of instruction

English

Number of ECTS credits

5

Mode of study

Not applicable.

Guarantor

doc. Ing. Václav Jirsík, CSc.

Department

Department of Control and Instrumentation (UAMT)

Learning outcomes of the course unit

Course graduate should be able to:
- explain the concept of artificial intelligence from the perspective of its application in technical equipment,
- explain the paradigm for artificial neural network: perceptron, multilayer neural network backpropagation learning, Kohonen self-organizing maps, Hopfield network, RCE neural network,
- discuss and verify the settings of individual parameters of the selected neural network,
- assess the scope of application of artificial neural network,
- explain the architecture and functionality of knowledge systéme,
- create a base of knowledge for expert system NPS32,
- choose the field of application of expert systéme,
- optical information processing devices applied artificial inteligence.

Prerequisites

The subject knowledge on the Bachelor´s degree level is requested.

Co-requisites

Not applicable.

Planned learning activities and teaching methods

Techning methods include lectures and computer laboratories. Students have to write a single project during the course.

Assesment methods and criteria linked to learning outcomes

Condition of petting full credit is absolut (100%) attendance in obligatory parts of lessons - the computer exercises and obtaining at least 15 points. Students are tested continuously and i tis possible to get maximum 20 points. The final written exam is rated by 70 points at maximum and the oral exam is rated by 10 points at maximum.

Course curriculum

1. Artificial intelligence - definition, history, area
2. Neuroscience - biological information system, neuron, brain, intelligence
3. Principles of computer vision
4. Artificial neural networks - definitions, paradigms
5. Perceptron, learning
6. Multilayer neural network with Backpropagation learning algorithm
7. Kohonen's self-organizing map, Hopfield network, RCE network
8. Convolutional neural network
9. Artificial Intelligence and practice
10. Representation of knowledge, problem solving
11. Expert Systems - definition, structure, knowledge base, application
12. Intelligent robot

Work placements

Not applicable.

Aims

The aim of this course is to provide students with a basic orientation in key algorithms and artificial inteligence, emphasis is placed on the field of artificial neural networks, knowledge systems and computer vision.

Specification of controlled education, way of implementation and compensation for absences

The computer exercises are compulsory, the properly excused missed computer exercises can be compensate.

Recommended optional programme components

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

Kasabov,N.K.: Foundations of Neural Networks, Fuzzy systems and Knowledge Engineering.The MIT Press,1996,ISBN 0-262-11212-4 (EN)
Schalkoff,R.J.:Artificial Neural Networks. The MIT Press,1997,ISBN 0-07-115554-6 (EN)
Sonka M., Hlavac V., Boyle R.: Image Processing, Analysis and Machine Vision. Thomson, 2008, ISBN 978-0-495-08252-1 (EN)

Recommended reading

Not applicable.

Classification of course in study plans

  • Programme EEKR-MN Master's

    branch MN-TIT , 1. year of study, winter semester, compulsory
    branch MN-KAM , 2. year of study, winter semester, compulsory
    branch MN-EEN , 2. year of study, winter semester, compulsory

Type of course unit

 

Lecture

39 hours, optionally

Teacher / Lecturer

doc. Ing. Václav Jirsík, CSc.

Syllabus

Artificial intelligence,definition,methods
Neural networks, biological neural units
Model of neurons and paradigm of neuron nets
Multilayer perceptrons, backpropagation algorithm, modified algorithms BP
Neural networks as associative memories, RCE neural network, Kohonen maps
Expert systems, principle, structure
Knowledge reprezentation, logic, production rules
Knowledge reprezentation, semantic nets, frames
Problem solving, type of problems, heuristic
Problem solving methods
Methods of inference
Speech recognition, processing, simulation and synthesis of speech
Speech recognition, methods of pattern recognition, using for instruments

Exercise in computer lab

26 hours, optionally

Teacher / Lecturer

doc. Ing. Václav Jirsík, CSc.

Syllabus

Matlab with Simulink
Backpropagation algorithm modelling 1
Backpropagation algorithm modelling 2
Dynamic system modelling by neural network
Sensitivity analysis of neural networks
Pattern recognition by neural networks
Expert systems application