BA Seminar 2 WF-FI-KGN-SLPE2

Diploma Seminar/Teaching Methods: Verbal and visual presentation of program content in the form of a PowerPoint presentation and demonstration of network operation and applications using specific examples for the preparation of theses/projects.

1. Invited seminar on the application of artificial neural networks and artificial intelligence in the broad sense;
2. Dynamic networks and their application in modeling and identification; Neural modeling; Dynamic neural models: the IIR filter model; Dynamic network architectures;
3. Neural networks in the identification of Wiener and Hammerstein systems; Wiener and Hammerstein system identification method: correlation methods, parametric regression methods, nonparametric regression methods, methods using gradient optimization techniques; Methods using neural models; Learning algorithms for neural models of Wiener and Hammerstein systems;
4. Neural networks in image processing and recognition; Examples of networks used for signal processing and recognition – images: silhouette recognition, license plate recognition;
5. Selected areas of biomedical applications of neural networks; Overview of thematic areas in which neural networks are most frequently used in biocybernetics and biomedical engineering; Classification of objects and signals; Signal clustering tasks; Neural models of complex biomedical and medical signals;
6. Neural methods for time series analysis and their possible applications in biomedical problems; Features of neural models; Limitations of the applicability of neural time series models; a structured neural time model; principles of fMRI image time series analysis;
7. Neural networks in modeling neuropsychological disorders: memory disorders, language disorders, Alzheimer's disease, mental illness; Attractor networks;
8. Neural methods for knowledge discovery in data: Logical rule extraction, rule optimization, rule fuzzification;
9. Neural networks and machine learning: an attempt at integration; Parameterization of similarity measures; Feature selection and scaling; Parameterization of functions weighing the influence of reference vectors; Mixing multiple models; Associative memory, pattern completion, and missing values;
10. Genetic and evolutionary algorithms; Mathematical foundations of algorithms; Application of genetic and evolutionary algorithms to neural networks;
11. Fuzzy and fuzzy-neural systems: fuzzy sets and fuzzy logic; Fuzzy systems; Fuzzy-neural systems: basic architectures of fuzzy-neural systems, training of fuzzy-neural systems;
12. Intelligent computing systems and artificial intelligence: concept, implementation, and application;
13. Hybrid methods in medical diagnostics: Data clustering algorithm; Fuzzy-neural rule base; Architecture and training of fuzzy-neural systems; Application to diagnosis in cardiology and oncology;
14. Project discussion;
15. Summary.

Syllabus developed by: Paweł Pęczkowski - course coordinator