Type of Credit: Elective
Credit(s)
Number of Students
This course presents general Bayesian principles and Bayesian computation techniques. It will cover empirical Bayes methods, Bayesian hierarchical models, Markov Chain Monte Carlo methods, and selected topics from Bayesian machine learning.
能力項目說明
Upon successful completion students should be able to formulate Bayesian models and use statistical software for data analysis.
| 教學週次Course Week | 彈性補充教學週次Flexible Supplemental Instruction Week | 彈性補充教學類別Flexible Supplemental Instruction Type |
|---|---|---|
Tentative schedule:
「課程主題」、「課程內容與指定閱讀」&「教學活動與作業」、「學習投入時數」
1. Introduction to Bayesian inference read slides & book 3 hrs
2. Empirical Bayes vs fully Bayes homework 1 3 hrs
3. Introduction to Markov Chain read slides & book 3 hrs
4. Markov Chain Monte Carlo homework 2 3 hrs
5. Markov Chain Monte Carlo read slides & book 3 hrs
6. Gibbs sampling homework 3 3 hrs
7. Gibbs sampling read slides & book 3 hrs
8. Convergence diagnosis and rstan homework 4 3 hrs
9. Exam (written)
10. Deterministic Bayesian appxorimation methods read slides & papers 3 hrs
11. Bayesian dynamic models homework 5 3 hrs
12. Mixture models read slides & papers 3 hrs
13. Exam (programming)
14. selected topics read papers 3 hrs
15. selected topics read papers 3 hrs
16. Project presentation read papers 3 hrs
17. Project presentation read papers 3 hrs
For more details on「課程內容與指定閱讀」、「教學活動與作業」, see handouts
1. homework + in-class exercise + discussion 30%
2. exam (written + programming) 40%
3. final project 30%
"Bayesian Data Analysis" by Andrew Gelman, John B. Carlin, Hal S. Stern, and Donald B. Rubin.
none