## Course Description

Probabilistic machine learning models are powerful, flexible tools for data analysis. This course focuses on models with latent variables, which simplify complex data sets by discovering hidden patterns in the data. By the end of the course you will be able to apply a variety of probabilistic models for text data, social networks, computational biology, and more. You will learn how to formulate probabilistic models to solve the data science tasks you care about, derive inference algorithms for these models using Bayesian inference techniques, and evaluate their performance. Topics will include topic models, stochastic blockmodels and latent space models for social networks, Bayesian inference, Markov chain Monte Carlo, mean field variational inference, and nonparametric Bayesian models. This course continues from CSE250A.

** Lecture time and venue: ** TuTh 6:30-7:50pm PETER 103.

** Instructor: ** James Foulds

** Instructor email: ** jfoulds at ucsd dot edu

** Instructor office hours: ** TuTh 5-6pm Atkinson Hall 4401

** Teaching assistant: ** Long Jin

** Teaching assistant office hours: ** Monday 4:30-5:30 EBU3B Room B275

** Teaching assistant email: ** longjin at eng dot ucsd dot edu

** Grading: ** Homeworks 25% (5 of them, 5% each), Group Project 35%, Final 35%, Participation 5%

** Piazza: ** Sign up for this course at piazza.com/ucsd/spring2016/cse291d

** Poll Everywhere: ** PollEv.com/jamesfoulds656

## Prerequisites

## Required Textbooks

## Syllabus

Lecture | Summary | Details | Assessment | Required reading | |

3/29/2016 | Tuesday | Overview, motivating examples, recap of directed graphical models | Motivating applications - computational social science, social networks, computational biology. Recap on d-separation/Bayes ball/explaining away | Blei, David M. (2014). Build, compute, critique, repeat: Data analysis with latent variable models. Annual Review of Statistics and Its Application, Sections 1-3 | |

3/31/2016 | Thursday | Bayesian inference | Monty hall problem, frequentist vs Bayes, MAP vs MLE, full posterior vs point estimates, posterior predictive | HW1 out | Read "What is Bayesian Analysis?" at bayesian.org (scroll down, and be sure to follow the link to bayesian.org/Bayes-Explained). Then read Murphy Ch 3.1 - 3.2 |

4/5/2016 | Tuesday | Generative models for discrete data | Conjugate priors, beta/Bernoulli, Dirichlet/multinomial, urn process interpretations, naive Bayes document model | Murphy Ch 3.3 - 3.5.2 | |

4/7/2016 | Thursday | Exponential families and generalized linear models | Exponential families, Bayesian inference for EFs, writing distributions in EF form, GLMs | Murphy Ch 9.1 - 9.3 | |

4/12/2016 | Tuesday | Monte Carlo methods | Importance sampling, rejection sampling, why they fail in high dimensions | MacKay Ch 29.1 - 29.3 | |

4/14/2016 | Thursday | Markov chain Monte Carlo | Gibbs sampling, Metropolis-Hastings | HW1 due, HW2 out | MacKay Ch 29.4 - 29.6 |

4/19/2016 | Tuesday | Mixture models revisited | Mixtures of Gaussians/connection to K-means, mixtures of multinomials, EM for mixtures of exponential family distributions, Bayesian inference | Project proposal due | Murphy Ch 11.1 - 11.3 |

4/21/2016 | Thursday | Latent linear models | Factor analysis, probabilistic PCA, ICA | Murphy Ch 12.1, 12.2.4 - 12.2.5, 12.6 - 12.6.1 | |

4/26/2016 | Tuesday | Hidden Markov Models revisited | Applications to part of speech tagging, Factorial HMMs, input/output HMMs, direct Gibbs sampling, Forward filtering backward sampling, collapsed MCMC | Murphy Ch 17.3 - 17.4.2, 17.4.5, Gao, J. and Johnson, M. (2008). A comparison of Bayesian estimators for unsupervised Hidden Markov Model POS taggers. EMNLP, Sections 1-2 (don't worry too much about the variational inference part, since we haven't covered this yet) | |

4/28/2016 | Thursday | Evaluating unsupervised models | Log-likelihood on held-out data, posterior predictive checks, correlation with metadata, human evaluation | HW2 due, HW3 out | Dave Blei's notes on posterior predictive checks |

5/3/2016 | Tuesday | Markov random fields | Separation criterion, Hammersley-Clifford theorem, Ising models, belief propagation | Murphy Ch 19, up to 19.4.1. Ch 20, up to 20.2.2 | |

5/5/2016 | Thursday | Statistical relational learning and probabilistic programming | Markov logic networks, probabilistic soft logic, Stan/winBUGS | Domingos, P., Kok, S., Lowd, D., Poon, H., Richardson, M., Singla, P. (2008). Markov Logic. In L. De Raedt, P. Frasconi, K. Kersting and S. Muggleton (eds.), Probabilistic Inductive Logic Programming, Sections 1, 4, and 7. | |

5/10/2016 | Tuesday | Variational inference | EM redux (Neal and Hinton-style), KL-divergence and evidence lower bound, mean-field updates | Project midterm progress report due | Murphy Ch 21 up to 21.3. My notes. |

5/12/2016 | Thursday | Variational inference (continued) | Examples - Gaussian, linear regression. VBEM | HW3 due, HW4 out | Murphy Ch 21.5 |

5/17/2016 | Tuesday | Topic models and mixed membership models | LSA/PLSA, Genetic admixtures, LDA, EM for LDA | Blei, David M. (2012). Probabilistic topic models. Communications of the ACM, 55(4), 77-84. | |

5/19/2016 | Thursday | Topic models (continued) | Collapsed Gibbs sampler, variational inference. Topic modeling with LDA extensions | Asuncion, A., Welling, M., Smyth, P., & Teh, Y. W. (2009). On smoothing and inference for topic models. UAI up to Section 3 | |

5/24/2016 | Tuesday | Social network models | P1, P2, exponential family random graph models | Goldenberg A., Zheng, A.X., Fienberg, S.E. and Airoldi, E.M. (2010). A Survey of Statistical Network Models. Foundations and Trends® in Machine Learning: Vol. 2: No. 2 Ch 3 up to 3.6, can skip 3.3 | |

5/26/2016 | Thursday | Social network models (continued) | Stochastic blockmodels, mixed membership stochastic blockmodels, latent space models | HW4 due, HW 5 out | Goldenberg A., Zheng, A.X., Fienberg, S.E. and Airoldi, E.M. (2010). A Survey of Statistical Network Models. Foundations and Trends® in Machine Learning: Vol. 2: No. 2 Ch 3.8 - 3.9 |

5/31/2016 | Tuesday | Models for computational biology | Profile HMMs for protein sequence alignment. Phylogenetic models and coalescent models | Murphy Ch 17.3.1 subsection on Profile HMMs. What is Phylogeny? article from the Tree of Life Web Project, hosted at the University of Arizona. Teh, Y. W., and D. M. Roy. (2007). Bayesian Agglomerative Clustering with Coalescents. NIPS up to Section 2. | |

6/2/2016 | Thursday | Nonparametric Bayesian models | Chinese Restaurant process, Dirichlet process, Dirichlet process mixtures, Indian buffet process | Gershman, S. and Blei, D.M. A tutorial on Bayesian nonparametric models. Journal of Mathematical Psychology, 56:1–12, 2012. | |

6/7/2016 | Tuesday | Final exam 7:00p-9:59p PETER 103 | |||

6/9/2016 | Thursday | HW 5 due, project report due |

The syllabus may be subject to change. The details column is only a guideline of the content likely to be covered, and the dates on which material is covered may shift.

## Homework and Project Policy

## Academic Integrity

UCSD and CSE's policies on academic integrity will be strictly enforced (see http://academicintegrity.ucsd.edu/ and http://cseweb.ucsd.edu/~elkan/honesty.html). In particular, all of your work must be your own. Any exceptions will result in a zero on the assessment in question, and may lead to further disciplinary action. While you may verbally discuss assignments with your peers, you may not copy or look at anyone else's written assignment work or code, or share your own solutions. Some relevant excerpts from UCSD's policies are:

## Family Educational Rights and Privacy Act (FERPA) Notice

Please note that as per federal law I am unable to discuss grades over email. If you wish to discuss grades, please come to my office hours.

## Campus Resources

- Black Resource Center: brc.ucsd.edu
- Cross Cultural Center: ccc.ucsd.edu
- LGBT Resource Center: lgbt.ucsd.edu
- Raza Resource Centro: raza.ucsd.edu
- Women's Center: women.ucsd.edu/