B 326 Geometric Machine Learning and Network Biology

This course aims to give a systematic and timely overview of the most critical developments and tools in network science and machine learning on graphs and their applications in various research areas with particular reference to LifeSciences. Network science and network biology are essential application areas where statistical and heuristic methods are important. Topics include biological and molecular networks, how to construct these from data, and how to analyze these such networks. Recent developments in geometric machine learning generalize convolutional operators to non-Euclidian geometries. Topics include graph neural networks, generative modeling techniques for non-euclidean geometries, operators on manifolds, message passing in graphs, symmetries and group theory, embedding techniques, and representation learning.


CS 220 and CS 229, linear algebra, and working knowledge of using python