Course Unit Code | Course Unit Title | Type of Course Unit | Year of Study | Semester | Number of ECTS Credits | İDR511 | Sosyal Bilimlerde Yapay Zeka Uygulamaları I | Elective | 1 | 1 | 6 |
|
Level of Course Unit |
Third Cycle |
Objectives of the Course |
The aim of the course is to enable students to: i) know basic neural network models and learning algorithms and ii) use artificial neural network models and related learning algorithms in signal processing and control applications. |
Name of Lecturer(s) |
Doç. Dr. Hakan PABUÇCU |
Learning Outcomes |
1 | Classify artificial neural network models and algorithms in terms of structure, usage and place of use, | 2 | Choose the appropriate neural network model and learning algorithm for an application | 3 | To be able to run learning algorithms effectively in a software environment | 4 | Be able to use artificial neural network models and learning algorithms in signal processing and control applications |
|
Mode of Delivery |
Normal Education |
Prerequisites and co-requisities |
None |
Recommended Optional Programme Components |
None |
Course Contents |
Artificial neural network architectures and learning algorithms. Multilayer sensor, networks with radial base function and support vector machines. Regression / function approach, classification and clustering. Artificial neural networks for signal processing, filtering and pattern recognition. Artificial neural networks for system diagnostics and control. |
Weekly Detailed Course Contents |
|
1 | Biological motivation. Historical view | | | 2 | Classification of artificial neural network models and learning algorithms | | | 3 | Adaptive linear element, least squares algorithm and convergence analysis | | | 4 | Discrete sensor and sensor learning rule | | | 5 | Multi-layer sensor, back propagation algorithm and types, convergence analysis, over learning | | | 6 | Radial based networks, design with input and input-output clustering | | | 7 | Mid-term exam | | | 8 | Support Vector Machines, Mercer Therory, Kernel representation, Lagrange mulitplier | | | 9 | Generalization, Vapnik-Chervonenkis dimension | | | 10 | Pattern recognition, feature extraction, size and data reduction with artificial neural networks | | | 11 | 1-dimensional signal processing with artificial neural networks | | | 12 | Image processing with artificial neural networks | | | 13 | Image processing | | | 14 | System identification with artificial neural networks | | |
|
Recommended or Required Reading |
Introduction. S. Haykin, Neural Networks and Learning Machines, Pearson Education, 3rd Ed., 2009, ISBN13 9780131293762 ISBN10 0131293761.
Lecture Notes. |
Planned Learning Activities and Teaching Methods |
|
Assessment Methods and Criteria | |
Midterm Examination | 1 | 100 | SUM | 100 | |
Final Examination | 1 | 100 | SUM | 100 | Term (or Year) Learning Activities | 30 | End Of Term (or Year) Learning Activities | 70 | SUM | 100 |
| Language of Instruction | Turkish | Work Placement(s) | None |
|
Workload Calculation |
|
Midterm Examination | 1 | 15 | 15 |
Final Examination | 1 | 40 | 40 |
Attending Lectures | 16 | 3 | 48 |
Self Study | 15 | 2 | 30 |
Homework | 5 | 10 | 50 |
|
Contribution of Learning Outcomes to Programme Outcomes |
LO1 | 3 | 4 | 4 | 4 | 5 | 5 | LO2 | 1 | 3 | 2 | 3 | 2 | 2 | LO3 | 2 | 5 | 5 | 2 | 3 | 3 | LO4 | 4 | 3 | 2 | 4 | 2 | 4 |
|
* Contribution Level : 1 Very low 2 Low 3 Medium 4 High 5 Very High |
|
|