Instructor: Yizhou Sun
Lecture times: T/R 10am-11:50am
Lecture location: BOELTER Hall 5264
The course aims to introduce probabilistic graphical models for structured data, where data points are no longer independent with each other, such as sequential data and graph/network data. The course will cover modeling, inference, and learning of state-of-the-art probabilistic graphical models, including Hidden Markov Model, Markov Random Field, Conditional Random Field, and Factor Graph. This course will also discuss recent trend on combining deep learning and PGM. Applications across different domains, such as text mining, medical domain, and social network analysis, will be discussed. The students are expected to read and present cutting edge research papers, as well as conduct a research oriented course project.
Part I: Lectures (5 weeks)
1. Introduction [slides] (0.5 lecture, Jan. 7)
2. Basic probabilistic models: Naive Bayes and Logistic Regression [slides] (1.5 lectures, Jan. 7, Jan. 9)
3. Hidden Markov Models [slides] (2 lectures, Jan. 14, Jan. 16)
4. Markov Random Fields [slides] (3 lectures, Jan. 21, Jan. 23, Jan. 28)
5. Conditional Random Fields [slides] (2 lectures, Jan. 30, Feb. 4)
6. Factor Graph [slides] (1 lecture, Feb. 7)
Part II: Paper Presentations (4-5 weeks, Feb. 11 - Mar. 12)
| Date | Paper | Presenters | Slides |
| Feb. 11 |
1. David M. Blei, Alp Kucukelbir, and Jon D. McAuliffe, “Variational Inference: A Review for Statisticians”, Journal of the American Statistical Association, Vol. 112 , Iss. 518, 2017. https://arxiv.org/abs/1601.00670 |
David Lee, Jiaxin Su, and Shuwen (Janet) Qiu | ppt; pdf |
| Feb. 13 |
2. Matthew Richardson and Pedro Domingos, "Makov Logic Networks", Machine Language, Vol. 62, Iss. 1-2, 2006. https://homes.cs.washington.edu/~pedrod/papers/mlj05.pdf |
Karen Quadros, Purit Punyawiwat, and Srishti Majumdar | ppt; pdf |
| Feb. 18 |
3. Meng Qu, Yoshua Bengio, Jian Tang, "GMNN: Graph Markov Neural Networks", ICML 2019. https://arxiv.org/abs/1905.06214 |
Cheng Peng, Chenlei Song, Qiyue Yao, and Shijia Hu | ppt; pdf |
| Feb. 20 |
4. Meng Qu and Jian Tang, "Probabilistic Logic Neural Networks for Reasoning", NeurIPS 2019. https://arxiv.org/abs/1906.08495 |
Zijie Huang, Roshni Iyer, Alex Wang | ppt; pdf |
| Feb. 25 |
5. Yuyu Zhang, Xinshi Chen, Yuan Yang, Arun Ramamurthy, Bo Li, Yuan Qi, Le Song, "Efficient Probabilistic Logic Reasoning with Graph Neural Networks", ICLR 2020. https://openreview.net/forum?id=rJg76kStwH |
Hengda Shi, Gaohong Liu, and Jian Weng | ppt; pdf |
| Feb. 27 |
6. KiJung Yoon, Renjie Liao, Yuwen Xiong, Lisa Zhang, Ethan Fetaya, Raquel Urtasun, Richard Zemel, Xaq Pitkow, "KiJung Yoon, Renjie Liao, Yuwen Xiong, Lisa Zhang, Ethan Fetaya, Raquel Urtasun, Richard Zemel, Xaq Pi, "Inference in probabilistic graphical models by Graph Neural Networkstkow", ICLR 2018 Workshop. https://arxiv.org/abs/1803.07710 |
Shihao Niu, Zhe Qu, Siqi Liu, and Jules Ahmar | ppt; pdf |
| Mar. 3 |
7. Matthew J. Johnson, David Duvenaud, Alexander B. Wiltschko, Sandeep R. Datta, Ryan P. Adams, "Composing graphical models with neural networks for structured representations and fast inference", NIPS 2016. https://arxiv.org/abs/1603.06277 |
Meet Taraviya, Joe Halabi, and Derek Xu | ppt; pdf |
Part III: Course Project Presentation (Mar. 12, 10-12pm)
Q & A
This term we will be using Piazza for class discussion. The system is highly catered to getting you help fast and efficiently from classmates and myself. Rather than emailing questions to the teaching staff, I encourage you to post your questions on Piazza.
Tips: Answering other students' questions will increase your participation score.
Find our class page at: http://piazza.com/ucla/winter2020/cs2493
"With its status as a world-class research institution, it is critical that the University uphold the highest standards of integrity both inside and outside the classroom. As a student and member of the UCLA community, you are expected to demonstrate integrity in all of your academic endeavors. Accordingly, when accusations of academic dishonesty occur, The Office of the Dean of Students is charged with investigating and adjudicating suspected violations. Academic dishonesty, includes, but is not limited to, cheating, fabrication, plagiarism, multiple submissions or facilitating academic misconduct."
For more information, please refer to the guidance.