Machine Learning Principles and Methods
General
- Course Code: 1802
- Semester: 8th
- Course Type: Scientific Area (SA)
- Course Category: Compulsory (CO)
- Scientific Field: Data Management - Artifial Inteligence (DMAI)
- Lectures: 4 hours/week
- ECTS units: 6
- Teching and exams language: Greek
- Recommended prerequisite courses: (1301) Probability Theory and Statistics, (1101) Mathematics Ι, (1201) Mathematics II
- Coordinator: Goulianas Konstantinos
- Instructors: Goulianas Konstantinos
Educational goals
The aim of this course is to introduce the student to the basic principles and problems of machine learning such as pattern recognition, value prediction, and clustering. The necessary mathematical background is given including the basic programming tools for the implementation of and application of the ML algorithms. With the successful completion of the course the student will be able to:
- Know the basic methods of Machine Learning and their field of application
- Understand the basic problem types where machine learning can be applied
- Analyze simple learning problems and apply the appropriate methods for their solution
- Implement the basic ML models using the appropriate programming tools
- Evaluate the performance of machine learning models
Course Contents
- Introduction to Machine Learning, basic concepts, the problems of pattern recognition regression, clustering and feature extraction
- Useful mathematical concepts from linear algebra, matrix theory, eigenvalue decomposition, probability theory, and optimization theory
- Generalization, the cross-validation method
- Introduction to Artificial Neural Networks, the linear neuron, the Perceptron, and ADALINE models
- Multi-Layer Neural Networks, the Back-Propagation learning algorithm
- Competitive Learning, Self-organizing networks
- Basic Recurrent models, associative memory, the Hopfield model
- Support Vector Machines, the concept of margin, linear and nonlinear kernels, support vector regression
- Basic clustering methods, the k-means algorithm
- Feature extraction
- Principal Component Analysis (PCA), Factor Analysis
Teaching Methods - Evaluation
Teaching Method
- Face to face lectures
- Optional programming exercises
Use of ICT means
- Use of the moodle.teithe.gr e-learning platform
Teaching Organization
Activity | Semester workload |
Lectures | 52 |
Individual study and analysis of literature | 128 |
Total | 180 |
Students evaluation
Final written exam with a combination of multiple-choice questions, short answer questions, and problem-solving questions.
Optional exercises
Recommended Bibliography
Recommended Bibliography through "Eudoxus"
- Κωνσταντίνος Διαμαντάρας, Δημήτρης Μπότσης, "Μηχανική Μάθηση", Εκδόσεις Κλειδάριθμος ΕΠΕ, Έκδοση: 1η/2019, ISBN: 978-960-461-995-5, Κωδικός Βιβλίου στον Εύδοξο: 86198212
- Κωνσταντίνος Διαμαντάρας, "Τεχνητά Νευρωνικά Δίκτυα", Εκδόσεις Κλειδάριθμος ΕΠΕ, Έκδοση: 1η/2007, ISBN: 978-960-461-080-8, Κωδικός Βιβλίου στον Εύδοξο: 13908
Complementary international bibliography
- Bishop, Christopher M., "Pattern recognition and machine learning", Springer, 2006.