EMÜ669 - PROBABILISTIC GRAPHICAL MODELS
| Course Name | Code | Semester | Theory (hours/week) |
Application (hours/week) |
Credit | ECTS |
|---|---|---|---|---|---|---|
| PROBABILISTIC GRAPHICAL MODELS | EMÜ669 | Any Semester/Year | 3 | 0 | 3 | 10 |
| Prequisites | ||||||
| Course language | Turkish | |||||
| Course type | Elective | |||||
| Mode of Delivery | Face-to-Face | |||||
| Learning and teaching strategies | Lecture Discussion Team/Group Work Drill and Practice Problem Solving Other: Lectures, drill and practice, problem solving, discussions, individual and group studies. | |||||
| Instructor (s) | To be determined by the department | |||||
| Course objective | The course aims to provide basic understanding of the properties of probabilistic graphical models including Bayesian networks and influence diagrams, the skills for building such models from expert knowledge, data or a combination of both, and evaluating them using sensitivity analysis and cross validation, and the ability to use them for risk analysis and decision support problems. | |||||
| Learning outcomes |
| |||||
| Course Content | Basic Properties of Bayesian Networks D-Separation Variable Elimination and Junction Tree Algorithms Dynamic Discretisation Algorithm Influence Diagrams Parameter Learning by MLE and Bayesian Learning Structure Learning Parameter and Evidence Sensitivity Cross-Validation Risk Analysis and Decision Support Models based on Bayesian networks and influence diagrams | |||||
| References | Fenton, Norman, and Martin Neil. Risk assessment and decision analysis with Bayesian networks. CRC Press, 2012. Kjaerulff, Uffe B., and Anders L. Madsen. Bayesian networks and influence diagrams. Springer, 2012. Koller, Daphne, and Friedman, Nir. Probabilisitic Graphical Models: Principles and Techniques. MIT Press, 2009. | |||||
Course outline weekly
| Weeks | Topics |
|---|---|
| Week 1 | Introduction to Risk Analysis and Probabilistic Graphical Models |
| Week 2 | Review of Probability Theory and Bayes Theorem |
| Week 3 | Introduction to Bayesian Networks |
| Week 4 | Graphical Properties of Bayesian Networks and D-Separation |
| Week 5 | Inference in Bayesian Networks: Variable Elimination and Junction Tree Algorithm |
| Week 6 | Eliciting the Structure of Bayesian Networks from Domain Experts |
| Week 7 | Learning the Parameters of Bayesian Networks from Data: MLE and Bayesian Learning |
| Week 8 | Midterm Exam |
| Week 9 | Learning the Structure of a Bayesian Networks: Constraint and Score Based Methods |
| Week 10 | Hybrid Bayesian Networks and Dynamic Discretisation Algorithm |
| Week 11 | Sensitivity Analysis |
| Week 12 | Cross-Validation |
| Week 13 | Influence Diagrams |
| Week 14 | Risk Analysis and Decision Support Applications of Bayesian networks and Influence diagrams |
| Week 15 | Study for Final Exam |
| Week 16 | Final Exam |
Assesment methods
| Course activities | Number | Percentage |
|---|---|---|
| Attendance | 0 | 0 |
| Laboratory | 0 | 0 |
| Application | 0 | 0 |
| Field activities | 0 | 0 |
| Specific practical training | 0 | 0 |
| Assignments | 3 | 10 |
| Presentation | 1 | 5 |
| Project | 1 | 10 |
| Seminar | 0 | 0 |
| Midterms | 1 | 25 |
| Final exam | 1 | 50 |
| Total | 100 | |
| Percentage of semester activities contributing grade succes | 6 | 50 |
| Percentage of final exam contributing grade succes | 1 | 50 |
| Total | 100 | |
WORKLOAD AND ECTS CALCULATION
| Activities | Number | Duration (hour) | Total Work Load |
|---|---|---|---|
| Course Duration (x14) | 14 | 3 | 42 |
| Laboratory | 0 | 0 | 0 |
| Application | 0 | 0 | 0 |
| Specific practical training | 0 | 0 | 0 |
| Field activities | 0 | 0 | 0 |
| Study Hours Out of Class (Preliminary work, reinforcement, ect) | 13 | 7 | 91 |
| Presentation / Seminar Preparation | 1 | 15 | 15 |
| Project | 1 | 50 | 50 |
| Homework assignment | 3 | 15 | 45 |
| Midterms (Study duration) | 1 | 22 | 22 |
| Final Exam (Study duration) | 1 | 35 | 35 |
| Total Workload | 34 | 147 | 300 |
Matrix Of The Course Learning Outcomes Versus Program Outcomes
| D.9. Key Learning Outcomes | Contrubition level* | ||||
|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | |
| 1. Reach the necessary knowledge and methods in engineering within the scope of advanced industrial engineering studies through scientific research and evaluate knowledge and methods and implement them. | X | ||||
| 2. Implement advanced analytical methods and modeling techniques to design processes, products and systems in an innovative and original way and improve them | X | ||||
| 3. Have the competency to plan, manage and monitor processes, products and systems. | X | ||||
| 4. Evaluate the data obtained from analysis of the processes, products and systems, complete limited or missing data through scientific methods, develop data driven solution approaches. | X | ||||
| 5. Develop original methods for the efficient integration of the scarce resources such as man, machine, and material, energy, capital and time to the systems and implement these. | X | ||||
| 6. Effectively utilize computer programming languages, computer software, information and communication technology to solve problems in the field of industrial engineering. | X | ||||
| 7. Report and present advanced studies, outcomes/results and the evaluations on the design, analysis, planning, monitoring and improvement of processes, products and systems. | X | ||||
| 8. Are aware of the professional responsibility, describe the technological, economic and environmental effects of the industrial engineering applications, work as an individual independently and as a team member having an understanding of the scientific ethical values, take responsibility and lead the team. | X | ||||
| 9. Are aware of the up-to-date engineering applications, follow the necessary literature for advanced researches, have the competency to reach knowledge in a foreign language, to quote and implement them. | X | ||||
*1 Lowest, 2 Low, 3 Average, 4 High, 5 Highest