“On the graph Laplacian and its applications in machine learning – Part 2: applications in machine learning”  by Gabriel Chen PhD, Mathematics & Statistics Department

This is the second of a two-talk series, in continuation with the first one given on February 8, 2024. In this talk we will start with a quick, brief review of the concepts and properties of the graph Laplaican matrix, so attending the first talk is not necessary for understanding this one. We will then focus on introducing the applications of the graph Laplacian to three areas of machine learning: nonlinear dimensionality reduction, spectral clustering, and graph convolutional nets. We will also touch upon some of the ongoing research on the scalability of spectral clustering and the design of graph convolutional nets. The talk is most accessible to people who have taken one semester of linear algebra; knowledge of graph concepts is helpful but not needed.