Instructor: Scott Linderman
TA: Xavier Gonzalez
Term: Spring 2023
Stanford University
Probabilistic modeling and inference of multivariate data. Topics may include multivariate Gaussian models, probabilistic graphical models, MCMC and variational Bayesian inference, dimensionality reduction, principal components, factor analysis, matrix completion, topic modeling, and state space models. Extensive work with data involving programming, ideally in Python.
Students should be comfortable with probability and statistics as well as multivariate calculus and linear algebra. This course will emphasize implementing models and algorithms, so coding proficiency is required.
- Time: Tuesday and Thursday, 10:30-11:50am
- Level: advanced undergrad and up
- Grading basis: credit or letter grade
- Office hours:
- Weds 4:30-5:30pm in Wu Tsai Neurosciences Instiute Room M252G (Scott)
- Tuesday 5-7pm location TBD (Xavier)
- Murphy. Probabilistic Machine Learning: Advanced Topics. MIT Press, 2023. link
You may also find the following references helpful:
- Bishop. Pattern recognition and machine learning. New York: Springer, 2006. link
- Gelman et al. Bayesian Data Analysis. Chapman and Hall, 2005. link
Week 1: Multivariate Normal Models and Conjugate Priors
Week 2: Hierarchical Models and Gibbs Sampling
Week 3: Continuous Latent Variable Models and HMC
Week 4: Mixture Models and EM
Week 5: Mixed Membership Models and Mean Field VI
Week 6: Variational Autoencoders and Fixed-Form VI
Week 7: State Space Models and Message Passing
Week 8: Bayesian Nonparametrics and more MCMC
Weeks 9 and 10: Research Topics in Probabilistic Machine Learning