Title: Hypergraph Co-Optimal Transport
Speaker: Samir Chowdhury (Bear Flag Robotics)
Speaker's URL: https://samirchowdhury.github.io
Abstract: Hypergraphs capture multi-way relationships in data, and they have consequently seen a number of applications in data analysis techniques that leverage higher-order network analysis. In this talk, we describe a framework for studying the space of hypergraphs using ingredients from optimal transport. First, we introduce a hypergraph distance based on the co-optimal transport framework of Redko et al. (2020) and present its theoretical properties. Next, we describe how our Hypergraph Co-Optimal Transport (HyperCOT) is computed in practice. Finally, we end with examples showcasing HyperCOT on various simulated and real-world datasets. This is joint work with Tom Needham (FSU), Ethan Semrad (FSU), Bei Wang (Utah), and Youjia Zhou (Utah). The associated arXiv preprint can be found on https://arxiv.org/abs/2112.03904.
URL associated with Seminar: https://tgda.osu.edu/activities/tdga-seminar/
