Advanced Topics in Machine Learning

Welcome to CS 290C, Advanced Topics in Machine Learning:
Advanced Analytics for Heterogeneous Information Networks

Instructor: Prof Lise Getoor
Course Time and Location: TTH 10:00-11:45 171 Soc Sci 2
Office hours: Th 2:00pm-4:00pm and by appointment, E2 347A

The course will survey a variety of machine learning approaches that are especially well-suited to inference and prediction in networks.   Topics will include: collective classification (predicting node labels in graphs), link prediction (predicting edges in graphs) and entity resolution (determining when two nodes refer to the same underlying entity).  In addition, we will cover a few of the probabilistic programming languages that are being developed to handle complex reasoning over relational and network data; in particular we will study probabilistic soft logic (developed by my group), and Markov Logic (developed by Pedro Domingo's group at UW), and others as time permits.   Depending on student interest and time, we may also spend a bit of time on learning causal models.

The class will be highly interactive, collaborative and fast paced (and hopefully fun!).   The bulk of the work will be a class project (which ideally relates to your research and results in a publication), and reading and presenting papers.   

Suggested background includes machine learning and probabilistic models, however course requirements will not be strictly enforced, just email me (, if you need a course enrollment code.   I expect that students will be coming to the course with a variety of backgrounds and will be adjusting the course to accommodate.

Course Format:

This is a seminar course. Each class will consist of presentations and discussion. Students will be required to do a class project for the course (60%) . A significant portion of the grade will be based on class participation, which includes paper presentations, contributions to the forum, and discussion (40%).

Instructors and Assistants