Logo

Mathematical Sciences Research Institute

Home » Community Detection in Sparse Random Hypergraphs

Seminar

Community Detection in Sparse Random Hypergraphs November 01, 2021 (02:00 PM PDT - 03:00 PM PDT)
Parent Program:
Location: MSRI: Simons Auditorium, Online/Virtual
Speaker(s) Yizhe Zhu (University of California, Irvine)
Description No Description
Video

Community Detection In Sparse Random Hypergraphs

Abstract/Media

To participate in this seminar, please register HERE.

The stochastic block model has been one of the most fruitful research topics in community detection and clustering. We consider the community detection problem in a sparse random tensor model called the hypergraph stochastic block model. Angelini et al. (2015) conjectured a threshold for detecting the community structure in this model, and we confirmed the positive part of the phase transition in the 2-block case. We introduced a matrix that counts self-avoiding walks on random hypergraphs, whose leading eigenvectors give a correlated reconstruction of the community. Based on joint work with Soumik Pal.

91870?type=thumb Community Detection in Sparse Random Hypergraphs 4.86 MB application/pdf

Community Detection In Sparse Random Hypergraphs