Course Description 

This course provides a concise and practical introduction to statistical inference and machine learning methods for analyzing network data, centered around three essential topics: community detection, network reconstruction, and graph representation learning. Within each of these topics, we will introduce the basic concepts then gradually move towards state-of-the-art techniques, along the way introducing key methods from statistical inference and machine learning that can be applied in more general contexts. We will illustrate the theory and algorithms with application examples from across the social and natural sciences. Students will gain hands-on experience through homework consisting of mathematical and/or programming exercises, as well as a final course project in the form of a research paper and presentation.  

Prerequisites 

A solid foundation in linear algebra, probability, and statistics will be important for understanding the course material. Such materials should at least be mastered at the level of MATH1853, but preferably at the levels of MATH2101/2102 for linear algebra and MATH3603 or STAT2601/2602 or STAT3902 for probability and statistics at HKU, or equivalent courses at other universities.  

Some exposure to Bayesian inference, data structures and algorithms, machine learning, and optimization is also recommended but not strictly necessary.   

If you’re curious about whether you would benefit from this course or your academic background is appropriate, contact the instructor for details.