What is a Beta-tree histogram?
Please note this event occurred in the past.
March 25, 2026 10:00 am - 11:00 am ET
LGRT 1685
Speaker: Qian Zhao, Umass Amherst
Abstract: Histograms are widely used to summarize and visualize data. However, multivariate histograms are difficult to construct because of the curse-of-dimensionality, which says that most of the histogram regions will be empty in higher dimensions. We will introduce a Beta-tree histogram, that uses data-adaptive partitioning to address this issue. The Beta-tree histogram possesses two properties: (1) with high confidence the distribution from which the data are generated is close to uniform on each rectangle in the histogram; and (2) we can provide simultaneous confidence intervals for the probabilities of each rectangle. We will show how to use the Beta tree histogram to visualize multivariate data and identify distinct populations in flow cytometry data.