Faculty of Information Technology, BUT

Course details

Soft Computing

SFC Acad. year 2019/2020 Winter semester 5 credits

Soft computing covers non-traditional technologies or approaches to solving hard real-world problems. Content of course, in accordance with meaning of its name, is as follow: Tolerance of imprecision and uncertainty as the main attributes of soft computing theories. Neural networks. Fuzzy logic. Nature inspired optimization algorithms. Probabilistic reasoning. Rough sets. Chaos.  Hybrid approaches (combinations of neural networks, fuzzy logic and genetic algorithms).

Guarantor

Deputy Guarantor

Language of instruction

Czech

Completion

Credit+Examination (written)

Time span

26 hrs lectures, 26 hrs projects

Assessment points

55 exam, 15 half-term test, 30 projects

Department

Lecturer

Instructor

Course Web Pages

Subject specific learning outcomes and competences

  • Students will acquaint with basic types of neural networks and with their applications.
  • Students will acquaint with fundamentals of theory of fuzzy sets and fuzzy logic including design of fuzzy controller.
  • Students will acquaint with nature-inspired optimization algorithms.
  • Students will acquaint with fundamentals of probability reasoning theory.
  • Students will acquaint with fundamentals of rouhg sets theory and with use of these sets for data mining.
  • Students will acquaint with fundamentals of chaos theory.

Generic learning outcomes and competences

  • Students will learn terminology in Soft-computing field both in Czech and in English languages.
  • Students awake the importance of tolerance of imprecision and uncertainty for design of robust and low-cost intelligent machines.

Learning objectives

To give students knowledge of soft-computing theories fundamentals, i.e. of fundamentals of non-traditional technologies and approaches to solving hard real-world problems.

Why is the course taught

By studying the subject, students will gain knowledge of working with vague, uncertain and incomplete information that is essential for successful intelligent system designs.

Prerequisite kwnowledge and skills

  • Programming in C++ or Java languages.
  • Basic knowledge of differential calculus and probability theory.

Study literature

  1. Kriesel, D.: A Brief Introduction to Neural Networks, 2005, http://www.dkriesel.com/en/science/neural_networks
  2. Munakata, T.: Fundamentals of the New Artificial Intelligence, Springer-Verlag New York, Inc., 2008. ISBN 978-1-84628-838-8
  3. Russel, S., Norvig, P.: Artificial Intelligence, Prentice-Hall, Inc., 1995, ISBN 0-13-360124-2, second edition 2003, ISBN 0-13-080302-2, third edition 2010, ISBN 0-13-604259-7

Fundamental literature

  1. Aliev,R.A, Aliev,R.R.: Soft Computing and its Application, World Scientific Publishing Co. Pte. Ltd., 2001, ISBN 981-02-4700-1
  2. Kriesel, D.: A Brief Introduction to Neural Networks, 2005, http://www.dkriesel.com/en/science/neural_networks
  3. Kruse, R., Borgelt, Ch., Braune, Ch., Mostaghim, S., Steinbrecher, M.: Computational Intelligence, Springer, second edition 2016, ISBN 978-1-4471-7296-3
  4. Mehrotra, K., Mohan, C., K., Ranka, S.: Elements of Artificial Neural Networks, The MIT Press, 1997, ISBN 0-262-13328-8
  5. Munakata, T.: Fundamentals of the New Artificial Intelligence, Springer-Verlag New York, Inc., 2008, ISBN 978-1-84628-838-8  
  6. Rutkowski, L.: Flexible Neuro-Fuzzy Systems, Kluwer Academic Publishers, 2004, ISBN 1-4020-8042-5
  7. Russel,S., Norvig,P.: Artificial Intelligence, Prentice-Hall, Inc., 1995, ISBN 0-13-360124-2, third edition 2010, ISBN 0-13-604259-7

Syllabus of lectures

  1. Introduction. Biological and artificial neuron, artificial neural networks.
  2. Acyclic and feedforward neural networks, backpropagation algorithm. 
  3. Neural networks with RBF neurons. Competitive networks.
  4. Neocognitron and convolutional neural networks.
  5. Recurrent neural networks (Hopfield networks, Boltzmann machine).
  6. Recurrent neural networks (LSTM, GRU).
  7. Genetic algorithms.
  8. Optimization algorithms inspired by nature.
  9. Fuzzy sets and fuzzy logic.
  10. Probabilistic reasoning, Bayesian networks.
  11. Rough sets.
  12. Chaos.
  13. Hybrid approaches (neural networks, fuzzy logic, genetic algorithms).

Syllabus - others, projects and individual work of students

Individual project - solving real-world problem (classification, optimization, association, controlling).

Progress assessment

  • Mid-term written examination - 15 points.
  • Project - 30 points.
  • Final written examination - 55 points; The minimal number of points necessary for successful clasification is 25 (otherwise, no points will be assigned).

Exam prerequisites

At least 20 points earned during semester (mid-term test and project).

Schedule

DayTypeWeeksRoomStartEndLect.grpGroupsInfo
Frilecturelectures G202 12:0013:50 1MIT 2MIT MBI MIN

Course inclusion in study plans

Back to top