(in Polish) Wstęp do sztucznych sieci neuronowych BD5020

Biological inspiration SSN, biological neuron, properties of SSN, historical background of SSN, artificial neuron and his action, activation functions, scaling data, action of neural network, regression and classification neural models, network design, network architecture: unidirectional networks, recurrent networks, Kohonen networks, networks of radial base function. Layered networks: numbers of network parameters, linear neural networks, multi-layered networks, training control algorithm, network learning algorithm, objective function, error back propagation algorithm, “overfitting”, measures of network error. Construction and operation of simulator Statistica Neural Network: automatic designer of network, data loading, result interpretation of network action, selection of the best model.

(in Polish) Rodzaj przedmiotu

(in Polish) Obowiązkowy

Course coordinators

Maria Sulewska

Learning outcomes

B3_W01 Doctoral student has advanced knowledge of basic character for scientific field and science discipline and science disciplines connected with performed research area.
B3_U01 Doctoral student is able to effectively acquire information connected with scientific different sources, also in foreign languages and achieve correct selection and interpretation of those information.
B3_U02 Doctoral student can, using own knowledge, make critical evaluation of the test results and other creative work-own and other authors-and their contribution in development of represented discipline; in particular, can evaluate suitability and possibility of result utilization of theoretical work in practice.
B3_K01 Understand and feel the need of continuing education, improvement of professional and personal competence, analysis of latest achievements connected with represented scientific discipline.

Assessment criteria

Laboratory assessment: Design and defence-min.3,0
Examination of lecture: Participation in checking of level knowledge discussion-min 3,0.

Bibliography

1. Tadeusiewicz R., Gonciarz T., Borowik B., Leper B.: Odkrywanie właściwości sieci neuronowych przy użyciu programów w języku C#. Wyd. PAU, Kraków 2007.
2. Masters T.: Sieci neuronowe w praktyce. Programowanie w języku C++. WN-T, Warszawa 196.
3. Duch W., Korbicz Jk., Rutkowski L., Tadeusiewicz R.: Sieci neuronowe. Tom 6: Biocybernetyka i inżynieria biomedyczna. Akad. Ofic. Wyd. Exit, Warszawa 2000.
4. Osowski S.: Sieci neuronowe do przetwarzania informacji Ofic. Wyd. Polit. Warsz., Warszawa 2006.
5. Stanisz A.: Przystępny kurs statystyki z zastosowaniem STATISTICA PL na przykładach z medycyny. Vol. 1, 2, 3. StatSoft, Kraków 2006, 2007, 2007.

Term 2013Z: