### 확률 그래피컬 모델 (Probabilistic Graphical Models)

- 강의실: 공학 x관 xxx호
- Textbooks
- D. Koller and N. Friedman, Probabilistic Graphical Models: Principles and Techniques, MIT Press, 2009 [book]

- References
- TensorFlow [link]
- CMU lecture on probabilistic graphical models [link]
- Stanford lecture on probabilistic graphical models [link]
- Caltech lecture on probabilistic graphical models [link]
- NYU lecture on probabilistic graphical models [link]
- UChigago lecture on probabilistic graphical models [link]
- Brown U lecture on probabilistic graphical models [link]

- Lecture 1: Introduction, Bayesian networks [pdf]
- Lecture 2: Undirected graphical models [pdf]
- Lecture 3: Undirected graphical models II [pdf]
- Undirected graphical models[pdf]

- Lecture 4: Exact inference [pdf]
- Variable Elimination [pdf]
- Complexity [pdf]
- Variable Elimination: Basic Ideas [pdf]
- Variable Elimination: Algorithm [pdf]

- Lecture 5: Exact inference II (Junction tree algorithm) [pdf]
- Belief propagation [pdf]
- Other slides
- [ref] Clique Trees [pdf]
- [ref] Message Passing: Sum-Product [pdf]
- [ref] Message Passing: Belief Update [pdf]

- Lecture 6: Exact inference III (Junction tree algorithm) [Message passing I] [Message passing II]
- Clique trees [pdf]
- [ref] Junction tree algorithm (by R. Nallapati) [pdf]

- Lecture 7: Inference as optimization: Message passing [pdf], Mean field approximation [pdf], Variational approximation [pdf]
- [ref] Variational inference [pdf]
- Application: Latent Dirichlet Allocation (LDA) [pdf]
- Exponential families [pdf]
- KKT condition [pdf]
- Additional material: Mean field approximation [pdf]

- Lecture 8: Inference as optimization: Loopy belief propagation [pdf][pdf]
- Lecture 9: Monte-Carlo methods for inference: [pdf] [pdf]
- Examples: [pdf][pdf][pdf]
- Sampling: Lecture note at U of Toronto [pdf]
- Markov Chain Monte Carlo (MCMC): NIPS 2015 tutorial[pdf]
- Markov Chain Monte Carlo (MCMC): Lecture note at CMU [pdf]
- Handbook of Markov Chain Monte Carlo [link]

- Lecture 10: MAP inference: [pdf]
- [ref] MAP and dual decomposition [pdf]

- Lecture 11: Learning in undirected graphical models: [pdf]
- Introduction to Learning [pdf]
- [ref] Learning Bayesian networks [pdf]
- [ref] Learning Markov networks [pdf]

- Lecture 12: Partially observed data - Parameter estimation: [pdf]
- Lecture 13: Structure learning: [pdf]
- Lecture 14: Topic models and Dirichlet processes: [pdf][pdf]
- Lecture 15: Markov logic network [pdf]
- [ref] Paper on Markov logic network [pdf]

- Lecture 16: Bayesian Nonparametrics in Document and Language Modeling [pdf]
- [ref] Hierarchical Dirichlet Processes [pdf]
- [ref] A tutorial on Bayesian nonparametric models [pdf]

- Lecture 17: Dual decomposition [pdf]

- Assignment 1: [pdf]
- Assignment 2: [pdf]
- Assignment 3: [pdf]