Fundamentals of Neural Networks and Fuzzy Logic

Lecturer:

Marcel Jiřina (gar.)

Tutor:

Marcel Jiřina (gar.), Petr Hošek

Supervisor:

Department of Biomedical Informatics

Synopsis:

Neural networks - history, biological and artificial neural networks (ANN). Utilizing of ANN for image processing. Model of neuron (node, unit), activation function. Principles of ANN. Self-organizing maps (SOM), U-matrix, LVQ classifier. Multilayer perceptron network (MLP), backpropagation. Basic learning/training of MLP and learning modifications. Structure optimization, pruning of ANN, selection of inputa data. Associative memories, Hopfiled's network, ART network. Applications of ANN in biomedical engineering. Utilizing of ANN in expert systems and for data compression. Fuzzy sets and fuzzy logic. Principle of extension. Fuzzy relations, cylindrical extension. Fuzzy numbers and arithmetic. Fuzzy implications. Approximate reasoning. Systems based on fuzzy rules, fuzzy inference. Logical conjunction, t-norms, s-norms. Examples for fuzzy modeling and image processing.

Requirements:

Basics of mathematics and Matlab.

Syllabus of lectures:

1. Introduction. Pre-processing of data and data formats. Introduction to ANN

2. Model of simple neuron (node, unit). Multilayer perceptron ANN (MLP). Structure of the network and algorithms of learning (backpropagation and other methods).

3. RBF ANN, Kohonen's SOM - structure, features

4. Associative memories, Hopfields's ANN, ART ANN. Applications of ANN

5. Fuzzy sets and fuzzy logic. Principle of extension. Fuzzy relations, cylindrical extension.

6. Fuzzy numbers and arithmetic. Fuzzy implications. Approximate reasoning.

7. Systems based on fuzzy rules, fuzzy inference. Logical conjunction, t-norms, s-norms. Examples for fuzzy modeling and image processing.

Syllabus of tutorials:

1. Modeling of simple neuron and one-layer perceptron in Matlab

2. Designing of RBF ANN for a classification task

3. Designing of Kohonen's SOM

4. Designing of Hopfield's ANN

5. Designing of fuzzy sets, operations with fuzzy sets

6. Fuzzification, Fuzzy inference, defuzzification

7. Application of fuzzy logic for fuzzy system design. Applications in Matlab.

Study Objective:

Provide students with basic overview of artificial neural networks and fuzzy logic.

Study materials:

[1] Bishop C. M.: Neural Networks for Pattern Recognition. Oxford University Press, NewYork, 1995.

[2] Fausett, L.: Fundamentals of Neural Networks. Prentice Hall, New York, 1994.

[3] Hassoun M. H.: Fundamentals of Artificial Neural Networks. The MIT Press, Cambridge, Massachusetts, London, 1995.

[4] Haykin, S.: Neural Networks: A Comprehensive Foundation. Macmillan Publishing, New York, 1994.

[5] Rojas R.: Neural Networks: A Systematic Introduction. Springer-Verlag, Berlín, Heidelberg, New York, 1996.

[6] Kazuo Tanaka, An Introduction to Fuzzy Logic for Practical Applications, Springer-Verlag, New York, 1996

[7] Timothy Ross, Fuzzy Logic with Engineering Applications, Mcgraw-Hill College, 2004 (1. vyd.), 2007 (2. vyd.)

Note:

Time-table for winter semester 2011/2012:

	06:00–08:0008:00–10:0010:00–12:0012:00–14:0014:00–16:0016:00–18:0018:00–20:0020:00–22:0022:00–24:00
Mon
Tue
Fri	roomKL:B-415 Jiřina M. 08:00–09:50 ODD WEEK (lecture parallel1) Kladno FBMI Počítačová učebnaroomKL:B-501 Hošek P. 16:00–17:50 EVEN WEEK (lecture parallel1 parallel nr.101) Kladno FBMI Počítačová učebna BMI
Thu
Fri

Time-table for summer semester 2011/2012:

Time-table is not available yet

The course is a part of the following study plans:

• Navazující magisterský studijní obor Přístroje a metody pro biomedicínu - prezenční (compulsory elective course)