Talks and presentations

Active Learning in Graph Based Semi-Supervised Learning

March 26, 2021

Talk, BYU Mathematics of Machine Learningm 2021,

Guest Lecturer for Dr. Jared Whitehead’s Mathematics of Machine Learning graduate level course at BYU. Presented my work on model change active learning for graph-based semi-supervised learning (SSL) as well as gave some advice on applying for Ph.D. programs in Applied Mathematics.

Active Learning in Graph Based Semi-Supervised Learning

March 01, 2021

Talk, SIAM CSE 2021,

SIAM CSE 2021, Minisymposium on Theory and Applications of Graph-Based Learning. Presented my work on model change active learning for graph-based semi-supervised learning (SSL), where we use the approximate change in the underlying SSL model as a measure of usefulness in the active learning process. This approximate change is efficiently done for a family of graph-based SSL models, using only a subset of the graph Laplacian’s eigenvalues and eigenvectors.

A Probabilistic Perspective on Link Prediction via Effective Resistances

March 17, 2017

Poster, BYU Spring Research Conference, 2017,

Won prize for best presentation of my session of the BYU Spring Research Conference. Presented research on using effective resistance for use in link prediction by viewing link prediction as a probabilistic problem wherein we view the current graph’s edge set as a realization of draws from an underlying probability distribution determined by a ground truth graph’s effective resistances.

Spectral Clustering in Directed Networks

March 19, 2016

Poster, BYU Spring Research Conference, 2016,

Won prize for best presentation of my session of the BYU Spring Research Conference. Presented research on estimating the latent number of clusters in directed networks for use in spectral clustering. The sequence of smallest eigenvalues of the associated graph Laplacian matrix that are real turned out to be a good estimator for the latent clustering structure.