EE5420 MACHINE LEARNING BY PROBABILISTIC MODELS
| Course Code: | 5675420 |
| METU Credit (Theoretical-Laboratory hours/week): | 3 (3.00 - 0.00) |
| ECTS Credit: | 8.0 |
| Department: | Electrical and Electronics Engineering |
| Language of Instruction: | English |
| Level of Study: | Graduate |
| Course Coordinator: | Prof.Dr. İLKAY ULUSOY |
| Offered Semester: | Spring Semesters. |
Course Objectives
This course is designed to familiarize M.S. and Ph.D. students with a general knowledge about uncertainity, belief, probability, various distribution models, graphical models, directed and undirected probabilistic graphical models, inference and learning in such models. With this course, students will be able to design probabilistic graphical models for their own problems and provide inference and learning specifically for their problems. They will becaome to be able to provide some theoretical and practical improvements in such models.
Course Content
Probability, graphs, Bayesian networks, Markov networks, temporal models, state observation models, Gaussian networks, exact inference, map inference and approximate inference (sampling) in these models, probability distributions, graph parameter learning with complete and incomplete data, graph structure learning by complete and incomplete data.